Longtime since the original post but I made some progress replacing the gear in this scraper that might be of help for others in the same situation.
I was starting with the gear remains in this condition and the information previously posted in the thread from 2016.
I did some preliminary calculations and eventually obtained all the bits needed to helical mill on the Cincinnati machine I now have. I came up with a sacrificial gear with the dimensions as follows:
No. Teeth-37
Module-0.7mm
Helix Angle-18 degrees
I only had a small piece of phenolic on hand to experiment with so I decided to model the gear in Fusion360 and have a someone at work use the 3d printer to make a prototype in PLA as a check. It seemed to mesh well. I even ran it under speed but without a load and it sounded OK and didn't instantly self-destruct. It got me thinking if there were any other 3d printer filaments that might be a good alternative to the phenolic of the original. I eventually settled on having one made in Nylon. It was about $12 delivered to my door so I ordered one while I set out to machine a phenolic version.
The scrap of phenolic I had was canvas based. In the end it might not be the best choice. I can't tell from the remains of the original but I think it might be linen based. I machined a couple gears from it and tried them out. Both eventually failed under load during use. The tooth form of both the machined examples was a bit loose compared to the 3d printed version. I think I need to refine my cutter a bit as I am using a single tooth fly cutter and the tooth profile is as best I can get ground by hand with nothing but a loupe to help magnify it.
In the interim the Nylon printed gear showed up so I tried it out. I've used it now for probably a couple hours total run time under load and it seems to be holding up very well:
I can't say it is the long term solution but it seems to work so far and the price was right. I do plan on getting some linen based phenolic and trying to machine a better version from that. Ultimately at the speed this gear operates at it would be best to have this hobbed but I don't have that capacity and not sure I want to invest the time to make one of the hobbyist work arounds to get that ability. Especially when the 3d printed solution looks as promising as it has.
I can only say this solution has worked for me on this scraper model to the limited use I have put it so far but I'd be glad to share the 3d .STL file of the gear for anyone who wants it. Not sure if I can attach that here and it might be easiest to post it to one of the 3d printer file sharing sites. Let me know if there is any interest.
Dick