Ok, I give up. 7 start thread

[alloy]

Uncle!!!

Been trying to make a speedometer gear with 7 and 9 start threads and have made over 30 test parts and am no closer now then when I started.

I'm using mastercam to program it and just trying the different options i mind boggling.

I've determined (as close as I can) that the thread crest is .115 apart, left hand, 38 degree angle with a .02 raduis at the tip. I've even had special inserts ground up to this.

The gears I have just don't match up with what I'm making. Mastercam is weird. At first it was at least making a 7 start thread, now it's stuck on 5 starts even though I've told it 7. Then seemingly at random it will miss a thread. It has the option to use canned cycyle, or manual using G32. The G32 wroks but makes for a 700 line long program. G76 doesn't work. And I've not be able to decipher G76 yet.

I'm not sure how to figure the pitch out. I tried using the space between the crest of the threads, and calculating that way. It would be nice if the thread was 1" wide and divide it by the number of threads, but 1t's .610 wide.

Right now I'd gladly pay to have it programmed, but I really don't know who would do that.

 

[Janderso]

You are a better man than I sir.
That looks like the drive gear.
I can’t imagine.

 

[JimDawson]

Don't know if this helps but the lead = 8.6956 TPI (1 / 0.115) or maybe 3mm lead as 0.115 works out to 2.921 mm

what's the OD and the single depth?

 

[Lo-Fi]

Are you trying to recreate an existing gear?

Helicals are hellishly difficult to measure, as the pitch is necessarily different as you move deeper into the tooth. Measuring at the outside diameter doesn't help much as they're defined at the pitch circle diameter, not the OD. I have a feeling this is where you're struggling based on your comments about measuring thread crests and the first photo (but could very well be wrong!).

Have you tried making the gears in CAD to sanity check that what you're feeding the CAM software is actually achievable?

 

[bakrch]

The pitch you are measuring is to the next helix, so the thread lead should be 7x the pitch.

 

[mmcmdl]

Why won't G76 work ? I did many multiple lead threads in the past and never had issues with the canned cycles .

If I remember correctly ( and my memory is far from perfect ) , your first G76 code would read P1020060 for a single lead thread . Change the 1 to a 7 and input your dimensions needed on the second line .

 

[alloy]

It is a drive gear for a speedometer. I buy the plastic ones for a pretty good price, but I also use steel ones and the supplier hits me for $30 each on them. His shipping is $7.95 each even if bought in quantity. But he's the only game in town for them.

I used the same for a lead number, it worked out to .115 to start, .125, and up to .160 feed rate. The closet was .125.

Yes I am trying to copy an existing part. There is no information on them anywhere.

I've not tried making it in CAD. I'm just not good enough to do that. I tried, but failed miserably.

Here is the code mastercam code using G76. The control hangs up on line 8.

N1 %
N2 O1206
N3 (7 START GEAR)
N4 G0 T0505
N5 G18
N6 G97 S400 M03
N7 G0 G54 X2.2 Z2.0204 M8
N8 G76 P040019 Q0 R0
N9 G76 X1.675 Z-.75 P625 Q31 R0. F.71942
N10 G0 W.1028
N11 G76 P040019 Q0 R0
N12 G76 X1.675 Z-.75 P625 Q31 R0. F.71942
N13 G0 W.1028
N14 G76 P040019 Q0 R0
N15 G76 X1.675 Z-.75 P625 Q31 R0. F.71942
N16 G0 W.1028
N17 G76 P040019 Q0 R0
N18 G76 X1.675 Z-.75 P625 Q31 R0. F.71942
N19 G0 W.1028
N20 G76 P040019 Q0 R0
N21 G76 X1.675 Z-.75 P625 Q31 R0. F.71942
N22 G0 W.1028
N23 G76 P040019 Q0 R0
N24 G76 X1.675 Z-.75 P625 Q31 R0. F.71942
N25 G0 W.1028
N26 G76 P040019 Q0 R0
N27 G76 X1.675 Z-.75 P625 Q31 R0. F.71942
N28 M9
N29 G28 U0. W0. M05
N30 M30
N31 %

 

[mmcmdl]

Alloy , it seems you're missing a numeral in the first P code , there should be 7 digits . Have you tried using the canned cycle ?

 

[alloy]

This is the code mastercam puts out. Just don't know enough to fix it.

Yes tried it myself manually, doesn't work.

Use G32 instead generated by the cam system.

 

[mmcmdl]

The machine doesn't like something in line 8 , and it doesn't look correct as I remember , but I always used the machines' canned cycle . This may be different .