What would be a reasonable cost for this one-off job?

I want to make a 127 tooth metric transposing gear for my lathe. I'm debating having one wired out (just profile and bore) vs 3d printing. If I provide the files (dxf, stp, etc, whatever is requested) what could I expect to pay for the wire route?

Gear details: 127t, 14 dp, approx OD 9.25", approx 5/8" thick

Material can be provided but paying shipping both ways would likely make that costly.

Hi Countryboy 19:
The best person I can think of to talk to about this project is Zahnrad Kopf who participates in this forum and is a gear specialist with both hobbing equipment and wire EDM equipment in his facility.

Wire cutting a gear like this might be turn out to be an expensive way to go about it, especially if he already has a hob to match your gear specs.
A 9" diameter gear has a path length of about 28 inches plus 127 times the flank length per tooth x 2 (two flanks per tooth), so at a rough guess without measuring the actual profile, you've got maybe close to 50 inches of cut that you have to run around at least 3 times at maybe an average of 0.120 inches per minute, so around 20 hours, so close to 2500 Canadian dollars just for the wire time, never mind setup and programming.

BTW, just for your information, wire cutting a gear from a turned blank is a royal PITA; I'm always much happier when I can cut gears out of a square big plate I can hang easily into the work tank.
I like to cut the bore too; it's the easy way to get everything as concentric as the wire EDM can make it, so add a bit for that operation too.

That's one EXPENSIVE gear, almost as nutso as what Rush Gear might charge you.

A gear can be hobbed from a turned blank in a fraction of that time, so unless your need is for something wildly unique for which a hob is unrealistic to get, my guess is that Zahnrad is going to see this as I do.

However, he's the expert so if he recommends wire cutting, then go with that, but it's not gonna be cheap.

Hopefully he sees this post and responds, if not, you may try to contact him directly.
His contact information is on his website...here's a link:
American Machine & Gear Works

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

Oh Yeah: don't even THINK about metal 3D printing a gear like this...you'll get a useless crappy blob, not a precision gear, and for a transposing train you need good precision if you want decent metric threads once you install this gear into the geartrain of your lathe.
You'll also bash the snot out of the mating gear, so resist the temptation to ignore this bit of advice unless you can live with the likelihood of having to replace the mating gear.
Don't waste your time with a plastic 3D printed gear either.

Honestly nothing more sophisticated than a cheap as in sub $20 Chinese hob, a suitable nylon blank and a Bridgeport + rotary table + excel list of the angular increments made me a similar 127 tooth gear in 14dp at 5/8" thick in nylon in a couple of hours. So long as each tooth is good enough position wise to mate with a good gear your golden, angular errors measuring sub 0.5 degrees have minimal effects on any cut threads by the time there transposed through the rest of the gear train!

Cutting it in nylon is easier and they run quieter.

Sent a pm earlier, but had another thought: Boston Gear says they stock change gears by one tooth increments to 128 teeth. 14.5 deg pressure angle only, though. Link here.

implmex said:

Don't waste your time with a plastic 3D printed gear either.

Click to expand...

Thanks for the advice. I don't know that I agree with all of your post but I do appreciate your post because it provokes critical thinking and brings some good advice.

That being said, many are using 3D printed transposing gears with good experiences. Do you have experience with one?

adama said:

Honestly nothing more sophisticated than a cheap as in sub $20 Chinese hob, a suitable nylon blank and a Bridgeport + rotary table + excel list of the angular increments made me a similar 127 tooth gear in 14dp at 5/8" thick in nylon in a couple of hours. So long as each tooth is good enough position wise to mate with a good gear your golden, angular errors measuring sub 0.5 degrees have minimal effects on any cut threads by the time there transposed through the rest of the gear train!

Cutting it in nylon is easier and they run quieter.

Click to expand...

That was my plan but the hobs/cutters seem hard to source for 14 DP, 20 degree PA with a tooth count that high for a sub $20 price. I'm seeing $100+. Do you recall where you sourced yours?

neilho said:

Sent a pm earlier, but had another thought: Boston Gear says they stock change gears by one tooth increments to 128 teeth. 14.5 deg pressure angle only, though. Link here.

Click to expand...

The problem is 14 DP is an oddball pitch. Leblond seemed fond of it while many others did not. Boston gear jumps from 12 DP to 16DP...

Look up this old tony he just did this for his metric lathe on youtube

Hi again CountryBoy19:
I am presuming that my sniffy dismissal of 3D printed gears for your application are the bits you don't agree with, and I understand how you could think that way if you've been exposed to anyone who has claimed success with transposing gears made like that.

In the very short term, with a hobby application, you might get away with it for a time, and if the gear train is used infrequently and on small threads it might even be for a long time.

However, if you're running the lathe to make money and you strip the teeth in the middle of a high value part you will not be a happy puppy.
Similarly, if you expect to make precision threads on your lathe, the precision of all the elements of the gear train helps you, and although adama makes the point that a nylon gear faked on the rotary table is good enough for his needs and the transposing errors are small, there's still a good counterpoint to be made about eliminating or reducing those errors where possible.

If that is truly the case for YOUR needs too, then a Cheapo B&S style involute cutter and a rotary table are a better solution in my opinion than a 3D printed gear, simply because of the better physical properties of the milled gear, but I would choose Delrin over Nylon for its better strength and better dimensional stability in conditions of varying humidity.
Nylon is hygroscopic to a much greater extent than Delrin is, so you may find the backlash in your gear changes over time with a nylon gear, especially a large diameter one.
Also if the nylon gear does not remain round as it absorbs water, you will have an additional problem maintaining proper mesh between the gears

If you can't find what you're looking for in a B&S cutter, you can fake one, and to do so, Ivan Law's book on gear cutting will help you, so I encourage you to get a copy and read it.
It's intended for the hobbyist and it's a great resource I use a lot for prototyping.

A last point;
It is not possible to hob a gear with a standard hob on a Bridgeport with a standard dividing head or rotary table.
The hobbing process requires the gear blank to rotate synchronously with the hob, and the hob to be presented to the blank such that the helix angle of the hob is parallel to the axis of the blank.(a hob has teeth arranged in the form of a thread with gashes along its length to create cutting edges, so it has a helix angle)

It is however, possible to invoke a variant of the Sunderland gear generating process in which the involute profile is faked using a stacked array of cutting teeth arranged like the teeth on a properly mating rack, and the blank is indexed tooth by tooth as it would be using the B&S cutters, but the tooth form of the cutter is simple (straight sided rack rather than involute) so you can make it at home.
A gazillion Youtube videos exist showing this process and many believe incorrectly that it's a hobbing process but it's not.


So take all of this for what it's worth...free advice over the internet.
The main point however with respect to your original inquiry, is that a long cut path on the wire EDM is going to be an expensive proposition, and when the path is broken up into many small segments with corners, the wire slows down a LOT, so this seemingly obvious way forward gets to be an expensive deal killer very quickly.

Given where you're coming from and to satisfy your needs, I'd abandon the wire EDM idea and focus on the alternatives available to you but I stand by my contention that a 3D printed gear is a crappy solution for anything other than the most casual hobby use and even then, it's a point of vulnerability to disaster easily overcome with better methods.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

I have made several 127 gears for different lathes, all were milled using a B&S involute cutter on a horizontal mill and dividing head in mild steel.
In this case I don't know what centre you need, keyway, splines etc otherwise I would send you a price. BTW as 127 is a prime number, you will probably not get it on a conventional dividing head, check before you start.
For this application the pressure angle is not too critical as you are able to control the meshing of the gear train, I would cut it at 14.5* anyway.
PM me if you need help.

Where to even begin.

Most of the time, people wanting to 3D print gears are hobbyists, not working for a living. Most of that time, money is the driving factor. Not accuracy. 3D printed gears are not accurate. Period. End of story. I've written about it on here before and actually done the work to investigate it and wind up with that conclusion based upon having done so and inspected them with actual gear inspection equipment.

Want to save pennies? Go for it and be happy you saved money. Don't try and tell anyone with experience that it's accurate, though. Of course, none of that addresses the realities of their efforts failing more oft than not. And that is most often in 9" and 10" bench top machines. Not a 14" - 16" floor model, prone to cutting more substantial loads.

That usually leads to the inevitable bartering, hedging, and acquiescence of using substitutes for the correct transposing gear, bearing less accurate ratios. The common argument that threads don't really need to be that accurate, or the lead screw is probably worn more than the ratio error, or threads only engage each other for short distances, or any other host of excuses one uses to satisfy the argument or half measure. Personally, I take objection to the automatic acceptance of such half measures because it then begs the question of where it stops. Why bother at all? Why not just cut the closest Imperial thread and use lapping compound? Why make the part to hold tight tolerances if one is going to half ass the thread? Why strive for any manner of quality in one's efforts, at all? No thank you. One may as well start right out the gate with a piece of shite threading die from their box store hardware aisle and call it quality craftsmanship.

My personal favorite is "free hobbing" which is neither free, nor actually hobbing. But I digress...

Just my personal opinion, but if one wants to cut correct threads, then one gets the proper equipment. In this case, the correct gears. Regardless of where they are sourced from. No one makes any real money on these kinds of things unless they make ( and sell ) them in quantities of thousands. Simple economy of scale manufacture at work, there. And I say that as someone that makes the kits and transposing gears for Monarchs and SouthBends. Yes, even the 13", 14", & 16" lathes.

I used to offer them widely and freely with happiness and glee, despite all the admonitions from friends and colleagues. It took a lot longer than expected for the good will to get beaten out of me, but they finally did it. I will not paint with so broad a bush as to say "No one"..., but the truth of the matter is that VERY FEW hobbyists and individuals are willing to pay realistic prices for a quality gear to be made in single quantities. And I've lost the will to be as accommodating as I once used to be, making them at a loss, or at the very best ( and extremely rare ) case of breaking even. When the shop's so busy, one simply cannot afford to accommodate the thought of doing something to break even or lose money for the sake of it and at the expense of the regular customer and their work.

Inevitably, this leads to the request to allow the person to cut the blanks themselves and send in to be Hobbed or Shaped. It's all so very simple and easy to do. Right? And it will save so much money. Ask the few people that have actually followed through with that plan to see how easy and simple that turned out to be. Again, the bottom line for so many is money. Cost. Everyone wants quality in single pieces. And almost to a person, they want it at hundred and thousand part quantities pricing. Now, if all you want is a moderately well made specimen, then by all means it is positively possible to accommodate. The issue that then presents itself is that I have yet to come across another shop owner that wants their name associated with mediocre work and results, regardless the price.

For what it's worth, we are actually Hobbing 14DP, 14.5°PA gears for SouthBend lathes this week. ( We can also WEDM gears when the customer desires it. )

Good luck. My personal recommendation is to purchase the approximately shaped form mill, cut the resultant mate 127 times, and live with the offending noisiness of the result, unless you actually do want something better.

Hope that helps.

About 20 years ago, I bought an orphan lathe- its a pre-production sample 18x60 from a taiwan factory. So ONE was all that was ever made. The subsequent production models were just different enough that there are only about 80% parts overlap.

And I got a screaming deal on it, and it had about 4 hours on it- BUT, in those 4 hours, somebody had managed to crash it, and destroy two phenolic "sacrificial" gears in the drive train. Without flying to Taiwan, and speaking fluent Taiwanese chinese, there was no way to buy a drop in replacement.

So I had a gear shop make me two Delrin gears for it, both similar size to what you want.

To answer your question, a "reasonable" price, in about 2001, for two custom gears was about $1200.

Adjusted for inflation, you might be able to get your gear for $800 to $1000 today- but most of the shops that used to do that work have been going out of business around here. Thats a reasonable price for a one off, custom made gear.

I dont know what Zahnrad would quote, if he was still quoting one offs, but if it was a grand, I would jump on it.

Ries said:

To answer your question, a "reasonable" price, in about 2001, for two custom gears was about $1200.

Click to expand...

Ouch. If it was teeth-only and I couldn't do it in under two hours, I'd be ashamed of myself.

That's still more than what he will think is "reasonable" but no wonder I never made the big bucks. I never had the balls to cheat people that bad. $1200, jesus. And people wonder why the US can't be competitive.

Ries said:

Adjusted for inflation, you might be able to get your gear for $800 to $1000 today- but most of the shops that used to do that work have been going out of business around here. Thats a reasonable price for a one off, custom made gear.

I dont know what Zahnrad would quote, if he was still quoting one offs, but if it was a grand, I would jump on it.

Click to expand...

No, it wouldn't be anywhere near that bad. Not even close.

EmanuelGoldstein said:

Ouch. If it was teeth-only and I couldn't do it in under two hours, I'd be ashamed of myself.
That's still more than what he will think is "reasonable" but no wonder I never made the big bucks. I never had the balls to cheat people that bad. $1200, jesus. And people wonder why the US can't be competitive.

Click to expand...

Meh. Most places won't charge nearly that. Rush is well known for exorbitant prices, but most smaller shops that I am acquainted with don't try to take advantage like that. But they're not going to give it away, either. We're plenty competitive. Just not with chinese sweat shops, toes swept in corners. Ironically, we get some work from Britain our cousins 'cross the pond simply for their not being able to find that many Gear houses left that can do good work at decent pricing. I don't know this for personal fact, but that's the excuse that three have given us, now.

Since you said that you want to make one in your original post get an equally divided 127 tooth gear or dividing plate. Right now on ebay there are 127 tooth gears for sale. There is a ring gear of 127 teeth for 37 bucks. free shipping. get and do a direct index with a gear blank and cutter (Travers has involute 14DP cutters) and presto a gear for almost nothing but your time. If you cut one stack a few blanks and make some to resell and recoup your cost. Get a cast iron lifting weight for a blank. Practice with a piece of plastic/phenolic.

My prices were based on Seattle pricing, from 2001. Seattle prices today would be much more.

But any gearmaker worth working with is going to have hundreds of thousands invested in equipment, probably an equal amount in tooling, and have overhead costs.
Out here on the west coast, real estate costs, labor, taxes, utilities, and insurance, mean that having a complete gear shop is really really expensive, and hourly costs are high.
I know people who have had custom gears made more recently in the NW, and those prices are market prices around here.

I had a friend who paid a grand for a custom drive gear for a big blacksmithing power hammer, some years ago. For something standard, like a SB or a Monarch gear, that was predictible, I can see less, but for reverse engineering and making from scratch, I am pretty inured to those high prices in the PNW.

EmanuelGoldstein said:

Ouch. If it was teeth-only and I couldn't do it in under two hours, I'd be ashamed of myself.

That's still more than what he will think is "reasonable" but no wonder I never made the big bucks. I never had the balls to cheat people that bad. $1200, jesus. And people wonder why the US can't be competitive.

Click to expand...

There is no CHEATING anyone if you present an estimate and produce quality work. I made that mistake for about ten year, from 15-25 I did all kinds of work for peanuts. I was always broke and pissed off. Then I wised up... I have had people say I'm too expensive and I just smile and let them go elsewhere. If I'm going to open my tool box or flip on the welder I'm going to make money.

if the OP was considering to 3d print the gear, I'm pretty sure the quality gear for 1200usd or even 1000usd probably isn't an option

getting the 40:1 dividing head to index for 127 teeth isn't a big problem either, a 200mm diameter circle of 127 holes would have around 5mm spacing between the holes, so a 2mm index hole pattern can easily be made, draw it up in cad, print on paper, punch transfer to the new plate, precision will be more than enough, then just index every 40 holes to get the 127 teeth

implmex said:

Hi again CountryBoy19:
I am presuming that my sniffy dismissal of 3D printed gears for your application are the bits you don't agree with, and I understand how you could think that way if you've been exposed to anyone who has claimed success with transposing gears made like that.

In the very short term, with a hobby application, you might get away with it for a time, and if the gear train is used infrequently and on small threads it might even be for a long time.

However, if you're running the lathe to make money and you strip the teeth in the middle of a high value part you will not be a happy puppy.
Similarly, if you expect to make precision threads on your lathe, the precision of all the elements of the gear train helps you, and although adama makes the point that a nylon gear faked on the rotary table is good enough for his needs and the transposing errors are small, there's still a good counterpoint to be made about eliminating or reducing those errors where possible.

If that is truly the case for YOUR needs too, then a Cheapo B&S style involute cutter and a rotary table are a better solution in my opinion than a 3D printed gear, simply because of the better physical properties of the milled gear, but I would choose Delrin over Nylon for its better strength and better dimensional stability in conditions of varying humidity.
Nylon is hygroscopic to a much greater extent than Delrin is, so you may find the backlash in your gear changes over time with a nylon gear, especially a large diameter one.
Also if the nylon gear does not remain round as it absorbs water, you will have an additional problem maintaining proper mesh between the gears

If you can't find what you're looking for in a B&S cutter, you can fake one, and to do so, Ivan Law's book on gear cutting will help you, so I encourage you to get a copy and read it.
It's intended for the hobbyist and it's a great resource I use a lot for prototyping.

A last point;
It is not possible to hob a gear with a standard hob on a Bridgeport with a standard dividing head or rotary table.
The hobbing process requires the gear blank to rotate synchronously with the hob, and the hob to be presented to the blank such that the helix angle of the hob is parallel to the axis of the blank.(a hob has teeth arranged in the form of a thread with gashes along its length to create cutting edges, so it has a helix angle)

It is however, possible to invoke a variant of the Sunderland gear generating process in which the involute profile is faked using a stacked array of cutting teeth arranged like the teeth on a properly mating rack, and the blank is indexed tooth by tooth as it would be using the B&S cutters, but the tooth form of the cutter is simple (straight sided rack rather than involute) so you can make it at home.
A gazillion Youtube videos exist showing this process and many believe incorrectly that it's a hobbing process but it's not.


So take all of this for what it's worth...free advice over the internet.
The main point however with respect to your original inquiry, is that a long cut path on the wire EDM is going to be an expensive proposition, and when the path is broken up into many small segments with corners, the wire slows down a LOT, so this seemingly obvious way forward gets to be an expensive deal killer very quickly.

Given where you're coming from and to satisfy your needs, I'd abandon the wire EDM idea and focus on the alternatives available to you but I stand by my contention that a 3D printed gear is a crappy solution for anything other than the most casual hobby use and even then, it's a point of vulnerability to disaster easily overcome with better methods.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

Click to expand...

There are 2 things I disagree with.

#1 The accuracy of 3D printed gears: I believe you may be assuming I intended to use a sub-$1k desktop homeowner printer. Having used a variety of them in the last half decade I would agree that they have horrible accuracy. Printer I intended to use is a $200k+ Stratasys that a buddy has. He can print several materials with a relatively high degree of accuracy. I haven't spoken specific materials with him yet but he says he has materials that will work well. I fully realize it won't be as accurate as a cut/hobbed/shaved gear. I'm willing to accept minor inaccuracies.

#2 I disagree with the level of effect an inaccurate gear tooth will have on a thread. A 127 tooth gear is 2.8 degrees per tooth. Let's be generous to your proposition and say that tooth could have a 10% error, or .28 degrees. Neglecting gearbox ratio, my lead screw is 8 tpi. A .28 degree rotational error on the leadscrew translates to a .00009" error in position of the thread. For all practical purposes that is immaterial to the vast majority of threading jobs. IIRC most leadscrew aren't even that accurate when new.

My intended use: I'm not a hobby guy, but I'm not full-time. I have a full-time job in engineering and the shop is primarily for my own use in maintaining equipment etc as well as prototyping designs etc. I do hired work to help keep the shop profitable but I don't get much metric work. Primary metric work would likely be personal gunsmithing on foreign guns as well as odd-ball job-shop jobs.

I have considered making this on my rotab many times but ultimately time is killing that proposition. I have a 2 month backlog of work in the shop, my full time job is asking me to put in more overtime, and I have newborn twins at home that are killing all my free time...

Hi yet again CountryBoy19:
Good points all, but there are things presumed in your hopes that a 3D printed gear will be adequate for your needs.

Some are:
Your buddy asserts he has materials that will be strong enough...clearly you are an engineer and can probably make that determination better than I, but are you SURE his contention is correct.

I have not been overwhelmed with admiration for the physical properties of 3D printed parts and I have used all varieties made with probably every process available so I do have some exposure even though I've never done a formal analysis to support my remarks.

Additionally, as you know, the pitch error is not the only way a gear can be out of spec; of far more concern to me is the development of an out of round condition whether immediately or over time as residual processing stresses subside.

As you can easily visualize, this can result in far greater errors than 10% pitch errors of individual teeth, and if the gear binds and fails catastrophically the whole discussion around such errors are moot anyway.

However, I did also point out that the potential problems inherent in a 3D print are easy to circumvent and I made and continue to make a case that since an easy method exists to make a superior gear, I personally would take it, if only for peace of mind.

Call me a morbid pessimist, but I cannot bring myself to have the kind of confidence in 3D printed parts, even glass filled or carbon fiber reinforced parts to see this as a thoroughly engineered solution that can simply be installed and then considered good forever more.

An unqualified assertion of material suitability for a purpose like this is fraught in my opinion... I don't even know the forces involved nor do I know what the environmental effects will be over time but I've done enough projects and seen enough failures to be wary of the level of confidence implied by your buddy's assertions.

So yeah...I'd be nervous every time I fired up the lathe.
Your mileage will vary with your own level of confidence of course.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

cammer said:

Look up this old tony he just did this for his metric lathe on youtube

Click to expand...

Looks like he did it mostly bc he could. His dislike of the plastic gears seemed more about the gear ratio voice the material.

Peter Colman said:

I have made several 127 gears for different lathes, all were milled using a B&S involute cutter on a horizontal mill and dividing head in mild steel.
In this case I don't know what centre you need, keyway, splines etc otherwise I would send you a price. BTW as 127 is a prime number, you will probably not get it on a conventional dividing head, check before you start.
For this application the pressure angle is not too critical as you are able to control the meshing of the gear train, I would cut it at 14.5* anyway.
PM me if you need help.

Click to expand...

Your PM box is full. It's just a plain center that I'll turn a bronze bush for.

implmex said:

Hi yet again CountryBoy19:
Good points all, but there are things presumed in your hopes that a 3D printed gear will be adequate for your needs.

Some are:
Your buddy asserts he has materials that will be strong enough...clearly you are an engineer and can probably make that determination better than I, but are you SURE his contention is correct.

I have not been overwhelmed with admiration for the physical properties of 3D printed parts and I have used all varieties made with probably every process available so I do have some exposure even though I've never done a formal analysis to support my remarks.

Additionally, as you know, the pitch error is not the only way a gear can be out of spec; of far more concern to me is the development of an out of round condition whether immediately or over time as residual processing stresses subside.

As you can easily visualize, this can result in far greater errors than 10% pitch errors of individual teeth, and if the gear binds and fails catastrophically the whole discussion around such errors are moot anyway.

However, I did also point out that the potential problems inherent in a 3D print are easy to circumvent and I made and continue to make a case that since an easy method exists to make a superior gear, I personally would take it, if only for peace of mind.

Call me a morbid pessimist, but I cannot bring myself to have the kind of confidence in 3D printed parts, even glass filled or carbon fiber reinforced parts to see this as a thoroughly engineered solution that can simply be installed and then considered good forever more.

An unqualified assertion of material suitability for a purpose like this is fraught in my opinion... I don't even know the forces involved nor do I know what the environmental effects will be over time but I've done enough projects and seen enough failures to be wary of the level of confidence implied by your buddy's assertions.

So yeah...I'd be nervous every time I fired up the lathe.
Your mileage will vary with your own level of confidence of course.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

Click to expand...

Once again, good points. I do work with 3D printed materials regularly. 2 years ago I wouldn't even be considering this. It's remarkable what just 2 yrs can do to the availability of better materials for additive manufacturing. The industry is growing so quickly now that the technology is improving faster than most can keep up with. There are now 3D printed parts flight-accepted in military aircraft (won'tsay which but it goes really fast and carries lots of high end armaments); my buddy is the one that prints them.

The out of round thought is very valid... I'll get more info on the tolerances that can be held in that regard.