Wire EDM and nitriding? 5

[724napier]

Hello all,

I know some of you guys have a lot of experience when it comes to manufacturing gears and splines. I am looking to get a little help/input on manufacturing an internal involute spline.

To produce the internal spline we plan to use a wire EDM as we do not have the proper equipment to broach it. My fear with using a wire EDM is the re-cast layer that it produces. The adapter is going to be made from nitralloy and the teeth will be nitrided. So I am concerned that the re-cast layer may produce a brittle layer that when nitrided becomes even more brittle. Is it possible to remove this re-cast layer through hand lapping with a male plug?

I was hoping someone with experience could shed some more light on the issue and possible ways of getting around it.

Thank you

 

[mfgenggear]

724Napier

there is more than one way to process this Job
all the information given is valuable information and is
well given. however these are my suggestions.
since we have not seen any drawings it's tough to say.
I manufacture aircraft/aerospace applications.

heat treat the material rough stock, Bar Stock
semi finish turn & mill
stress relieve 50 degrees below the tempering temp.
wire edm spline. hold low limit to allow for stock removal of .0005 max per surface.
request that re-cast layer is not more .0002 per surface.
a very light bead blast spline area
deburr. pre-nitride is a must.
2 stage Nitride
chemical etch to remove white layer ( I do this all the time (speak with the heat treater) simple and easy, use gages to make sure it does not blow over if available)
NDT to insure white layer is removed.
inspect
finish grind if required for bearing surfaces.
stress relieve after grinding
mag particle inspect
final Inspect

thats it

HTH

Mfgeng

 

[mfgenggear]

forgot to add do a bake after chem etch. but the heat treater knows that.

 

[tbuelna]

I would consider the application semi-critical. It's going to be connecting a turbine engine to a dyno. Having all of these processes involved is making it a lot of trouble.

Whoa there! I would not consider a turboshaft engine output shaft spline to be a "semi-critical" application. They are usually highly engineered splines, and the OEM typically provides very specific requirements for the spline material properties, fits/tolerances, and operating environment in terms of installation misalignments, lubrication, dynamic balance, etc. The spline joints of turboshaft outputs usually use crowned external teeth which helps minimize radial loads on the engine shaft bearings due to installation or operating misalignments. These crowned splines have high flank contact stresses, so they typically use case hardened teeth. Since you are nitriding the internal spline teeth of your adapter, I'm assuming that the external teeth of the mating part are probably crowned. The crowned spline joints used on turboshaft outputs typically are major diameter fit type splines. The major diameter fit constraint minimizes radial movements between the mating halves of the spline joint from dynamic effects at high rpm.

Since this application is for a turboshaft engine output, I'd highly recommend shaping the spline rather than using wire EDM. I don't think you will obtain the accuracy this particular internal spline requires using wire EDM. If you would provide more specific details about the spline (ideally the ICD from the engine OEM), I'd be happy to provide more assistance.

Best regards,
Terry

 

[gearcutter]

tbuelna - you raise some very important points. Thanks for putting us back on-track.

FYI - Given the correct geometry/programming is used; wire EDM is of such high accuracy that parts manufactured this way can be used as gauges.

Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

http://www.aussieweb.com.au/email.aspx?id=1194181

 

[724napier]

Tbuelna,

You are correct that it is critical, but not flight critical. The male spline is on the engine dyno not on the turbine. It mates to the turbine engine by four bolts. The male shaft on the dyno is made from 15-5 PH and is not crowned. There is currently another shaft being used to connect the dyno to the turbine. It is made from nitralloy and is a fillet root side fit. There's been no issues with it and it has many hours on it. I am out of town at the moment, so that is all the info I have off the top of my head. Thanks for your help. You to gearcutter.

 

[tbuelna]

gearcutter-

I agree that the profile of a wire EDM surface by itself can be extremely accurate, but the problem with using wire EDM for an application like this is ensuring that the finished wire EDM surface is accurately located relative to other critical features of the part. Where I work we have a very nice Mitsubishi wire EDM machine, and one issue we have is how to accurately locate the start of an internal cut (like an internal spline) relative to the part after feeding the wire through the start hole.

 

[mfgenggear]

Terry

the main concern with EDM many companies don't allow it. far as accuracy it has been my experience there has been no issues holding P.D. runout to the parts datums.
the other problem from what have seen is the creation of the spline geometry. it has to be exact. normally it is best to very with gages or a mating part.

724napier

Gear cutter is correct that if plasma nitride is used. if done correctly there will not be any white layer. however hard tooling has to be used for masking. which gets more difficult. as whole EDM is a very reliable source of making splines. but you need to start 3 parts to make 1. because you need a part that they can use as a setup to verify the programing. and other tooling issues.

HTH

 

[tbuelna]

mfgenggear-

I agree that it is possible to produce an accurate internal spline using a good quality wire EDM machine. However, I would argue that you can usually get a better quality internal spline with less trouble using a conventional shaping process. You can usually find an aerospace gear vendor that has existing tooling for the splines used on aircraft drivetrains, so non-recurring costs should not be an issue. Personally, if given a choice between using wire EDM or shaping to produce this spline, I would choose shaping every time.

I would also agree that the best way to inspect splines is with master gauges. For most of the aircraft parts with splines that I have designed a set of go/no-go gauges was typically made, primarily because the parts with the mating halves of the spline joint were manufactured by different vendors. So the only practical way to ensure the mating halves of the spline had a proper fit was a set of master gauges that we could send to each manufacturer to validate their parts.

 

[mfgenggear]

Terry

hahahaha yes we could go back and forth on this one. shaping is a better route to go because it is assured there is the correct spline geometry. however when splines have small numbers of teeth, and are fine pitch, and are less than 1.00 Pitch Diameter.
there is a tendency for the cutter to deflect badly. dependending on the quantity of parts if it's a blind hole I go with punch broach. if it's a through hole I go with with a pull broach. if it's only 1-2 parts I will use wire edm or sinker if it's blind hole & if it's allowed by the customer to use ether. I will shape it if that is my last resort. and yes it does work. but remember it has to be tooled. the cutter has to have the correct whole depth or D+F, the cutter has to be the correct number of teeth so it will not side trim. it has to have the correct length so it will clear the tooling and the part.

plus depending the configuration if off the self tooling does not work it has to have a special pot to hold it. see it starts to add up.so now we are specking of dollars for one part. so if it is acceptable to wire cut by the customer it is an another tool in the bag. hahahaha

Yes gages are usually a must for mass production but for one or two parts a mating part will work. it just has to fit. and a 2 or 3 between wire inspection check is sufficient to make sure it's in the correct size. I can and do design spline gages. when it is necessary. or I purchase them from another spline gage manufacture.

the gears and spline I work with are and can be master gear quality or master spline quality from the old spec AGMA Q15 and even tighter.

sorry for the long post but just given my 2 cent.

Best Regards

MfgEng

 

[tbuelna]

hahahaha yes we could go back and forth on this one

But isn't the purpose of this forum to provide a place where gear nerds like you and I can endlessly debate an insignificant topic like this?

I'm pretty sure we lost 724napier long ago, but I hope we provided some useful advice before he checked out!

 

[724napier]

tbuelna,

I am still here. I have just had a hectic week and haven't gotten much time to respond. I am waiting to hear back from some vendors on having the spline shaped, since we do not currently have a shaper only an EDM. I have been soaking all the advice up from you mfgenggear and gearcutter. I actually had some more questions for you guys as well.

-When is it worth it to use a VAC-ARC material vs. regular air melt?
-As far as Go/No-Go gages, do you recommend making them on an EDM or is it better to buy a precision ground set? This is only a prototype part and will be very low production.
-Also in reference to the Go/No-Go gages. Is the Go gage typically made with the minimum effective circ space width while the No-Go is made to the max effective? It seems like it would be better to have more than just these two gages.

Thanks again to all you guys for the help. It's pretty amazing how deep this topic can get.

 

[724napier]

Also,

When using an EDM to wire cut an internal spline. Do you typically set the circular space width between the min/max effective or the min/max actual?

 

[mfgenggear]

724 Napier

Vac Arc is worth the extra cost.

Buy the set of gages gages that are certified.
For an Internal Spline the MIn Effective is the Go Gage
the max actual is the No Go Gage. is what I recommend. ANSI B92.1-1996.

Use the min actual for the EDM programing then offset to allow for the overburn
during the programing. use the gages to verify during cut.
and after heat treat.

 

[tbuelna]

724napier-

Unless you know the pedigree of the external spline your adapter will mate with, and/or have detailed documentation describing the interface requirements of your adapter spline, then as mfgenggear suggested for just two or three parts it would be best to simply match your adapter spline to the existing shaft spline. Having a certified pair of go no-go spline plug gauges made might cost $1000 or more, and it doesn't sound like you have the budget for this.

As for the cost/benefit of using vacuum melt quality materials, for the small number of pieces involved the higher raw material cost will not be a significant factor in the overall cost of the parts, so it is definitely worthwhile. Vacuum melt quality steels are very beneficial where fatigue life or reliability (fracture control) are concerns.

Good luck with your project.
Terry

 

[724napier]

Thanks mfgenggear and tbuelna, now if I could only find a supplier with vacuum melt material in stock I feel like I will be in good shape.