Help: Predicting Size Change for 17-4 After Heat Treating 1

[aeromatt]

Hello Eng-tips Gurus,

I should have wrote this prior to sending my parts in for heat treating and probably before bidding the job. I have a part to make that is machined from two pieces of 17-4 round stock. Part one is 1.63 dia x .38 long and Part 2 is .75 dia x 4.75. Part two is machined down to .19 for 3" then left at 3/4" for .625" and finally 1/2" for 1.13. There is a .25 dia hole down .75 in the 1/2" dia end. The narrow end of Part 2 is inserted in a hole at the center of Part 1 and welded.

If anyone is still with me I'm amazed. I have to bring this weldment up to H1150 which includes solution annealing for the weld and aging to H1150. I need to hold the 3/4" dimension from Part 2 to .7495-.7500 and the 1/2" dimension to .4996-.5000. Really I can't imagine why the tolerances are so tight but the question is can I hope to meet those tolerances if I were to try again. These are the only tight tolerances on the part and the only thing out of conformance after heat treating. For my first attempt I started out in tolerance on the high side before heat treatment. The 1/2" dia section with the hole down the center shrunk symmetrically about .001-.002" in diameter. The solid section at 3/4" dia ended up getting egg shaped about .002-.003" with the desired size somewhere in the middle.

I would machine it after heat treating but have not figured out a way to hold on to it. It might be the only way with these tolerances.

I bought the metal from mcmaster, that could be part of the problem.

Does anyone have any links or references for diametrical size change predictions for 17-4? Any advice on this matter would be appreciated.

Matt

 

[redpicker]

Buy material in solution treated condition.
Machine to size.
Weld as required.
Age at 1150F.

Why won't this work?

 

[aeromatt]

Thanks Redpicker. Isn't the solution annealing needed after welding to restore the 17-4 prior to heat treat. I think most of the shrinkage is during the aging process so I would still have to make my best guess at how much shrinkage will occur. The part that bothers me the most is the one section that became egg shaped after the process.

 

[EdStainless]

If you don't anneal after welding you really get the correct properties.
Now if the welds are not part of the critical load path then don't worry about annealing.

My guess is that your problem lies with the annealing.
If you are going to anneal after welding then you should anneal before also, it looks like your material wasn't uniform, and hence the distortion. I would stick to the low side of the annealing temp 1850-1900F.
The hole should get smaller when you age. There will be a volumetric contraction of 0.001"/inch.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[dgallup]

While the above may help, I'd really be surprised if you could hold .0004 tolerance after heat treatment. I would plan on leaving some extra stock and doing a finish grind in the 1150 condition.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[redpicker]

Yes, ideally you would solution anneal after welding, but sometimes we have to make do with less than ideal.

The structure of the weld metal and HAZ when you start with solution-annealed material is going to be a mixture of fully solution treated martensite and overaged precipitated martensite. The regions that are solution treated will age to H1150 during the aging treatment. The regions that are overaged (due the the heat from welding) are not going to have properties that are all that different than the aged H1150. Because the welded region is not that large, there isn't going to be all that much aging going on, regardless. If you were using either H900 or DH1150 (used for H2S applications) I don't think I'd try it. Ed is correct (at least, I assume he meant "If you don't anneal after welding you really won't get the correct properties"), but at the H1150 level with parts of this size, I don't think the differences are going to be much.

Yes, as Ed mentioned, there is a volumetric contraction on aging. In my experience, it is very predictable and uniform. I'd suspect the egg-shape resulted from non-uniform cooling during the annealing.

 

[aeromatt]

Thanks for the responses Ed and D. Ed, are you saying the annealed condition of the 17-4 from mcmaster is not good enough and that could have caused the egg shape after heat treat.

D, I agree, the tight tolerance and small quantity order leads me to think I need to plan for a post heat treat machining to size.

Thanks again everyone.

 

[aeromatt]

Thanks Redpicker. How can ensure uniform healing? Is there something the heat treater should be doing to provide uniform cooling?

 

[aeromatt]

Holy cow I was distracted in the last post. I meant to say, how can I ensure uniform cooling...?

 

[EdStainless]

When you cool this material from anneal it needs to be cooled reasonably quickly, but not a hard quench. The heattreater will cool as fast as his equipment allows to get faster turnaround. Having him back off on the cooling rate and perhaps changing how the parts are loaded.
Holding 1/2 thou after aging will be very difficult.
This stuff may have been batch annealed in a bundle so uniformity may not be that great.

Sorry, yes if you don't anneal the welds the properties are bit off. Fine from the mechanical stand point (you just loose a little toughness) but they can be very unreliable when it comes to holding tolerance and straightness.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[MagBen]

I bet the distortion was from annealing/solution, the contraction from aging. some fast cool after solution is needed to get a full martensitic structure. Since your diameter is relatively small, you may be able to decrease the cool rate a bit. The contraction was mainly becasue of austentic reversion, 1150H has the highest shrinkage about .001 per inch. H900 has the smallest, but still having .0005'/ in.

If the size change is your main concern, I recommend buying solutioned + aged material (but not overaged, say H1075). This status is weldable and machinable.

When the main ocncern is your egg shaped portion, you can buy solutioned condition, the alloy can be directly aged to the desired strength level after welded.

For the best combination of strength + ductility + corrosion, you do need to anneal + aging after welded. Then you have to live with the distortion and contraction.

 

[tbuelna]

I don't think you will be able to hold a .0004" tolerance on the 1/2" diameter near the weld joint after welding. And things won't improve in this area after heat treating. To hold this tight of a tolerance you'll need to finish grind the surface after welding and heat treating.

 

[aeromatt]

Thanks everyone for generously sharing all this information. I should have come here first.

Matt

 

[tbuelna]

aeromatt- If you have any questions like this in the future, MMPDS-01 (or MIL-HDBK-5) is an excellent technical resource and is free to download. For your reference here is what MMPDS-01 sec 2.6.9.0 has to say about the subject:

"Manufacturing Considerations — 17-4PH is readily forged, machined, welded, and brazed. Machining requires the same precautions as the austenitic stainless steels except that work-hardening is not a problem. Best machinability is exhibited by Conditions H1150 and H1150M. A dimensional contraction of 0.0004 to 0.0006 and 0.0008 to 0.0010 in./in. occurs upon hardening to the H900 and H1150 conditions, respectively.This fact should be considered before finish machining prior to aging treatment. When permanent deformation is performed, such as cold straightening of hardened parts, reaging is recommended to minimize internal stresses. Alloy 17-4PH can be fusion welded with any of the normal processes using 17-4PH filler metal without preheat. For details up to ½-inch thickness, Condition A is satisfactory prior to welding, but for heavy sections,an overaged condition (H1150) is recommended to preclude cracking. After welding, weldments should be aged or solution treated and aged."

 

[WKTaylor]

Aeromat...

Back-up to statement by tbuelna.

From ASMH Hdbk Code 1501 17-4PH [1995]...

1.5.2 During the aging treatment a predictable dimensional
change occurs, ranging from contractions of 0.0004-
0.0006 in./in. for the H900 treatment to 0.0008-0.0010
in./in. for the H1150 treatment. (Refs. 20, 21, 23)

References.
20. “Armco 17-4PH Precipitation-Hardening Stainless Steel
Sheets and Strip,” Armco Steel Corp., S-96,
LA-6772 (August 1972).

21. “Armco 17-4PH Precipitation-Hardening Stainless Steel
Sheets and Strip,” Armco Steel Corp., S-6c,
LA-10273 (January 1974).

23. “Armco 17-4PH Stainless Steel Bar and Wire,” Armco
Steel Corp., LA-3575 (April 1975).

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true.
o For those who believe, no proof is required; for those who cannot believe, no proof is possible.
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion"]
o Learn the rules like a pro, so you can break them like an artist. [Picasso]

 

[WKTaylor]

Weird note [war-story] Relative to this issue of thermal Precipitation HT inducing dimensional changes in PHT CRES and super alloys.

~15-years ago…

We had a drawing requirement for (2) 1.0000"-Dia I-718 Shear PINs machined to a very high tolerance to be installed in blind holes in a major steel fitting assembly. These PINs were to be installed bare into [bare] low alloy steel holes with ONLY corrosion preventative compound [CPC]. EVERY nerve in my body said this was absolutely a bad idea: high nickel alloy pin into a high HT LA steelhole with a non-chromated non-sealing/temporary corrosion protective coating was a terrible idea from a corrosion perspective. This was backed up by several documents [testing and service history].

I explored application of cadmium plating with a post-plating chromate treatment; then realized that the PINs probably needed to be ground undersized to a ‘before plating’ dimension… to allow plating thickness build-up to the final dimension.

So I took a quick-look at the drawing for 'any-other-issues'. Lo-and-behold, the material, AMS5662, came in the SHT condition and had to be precipitation hardened [PH] to (1) of (3) conditions: (1) per AMS2774/MIL-HDBK-5H or (2) per a company HT spec. When I reviewed the PIN drawing top-bottom, front-back, It dawned-on me that neither the raw stock nor the rough-machined PINs had a PHT step. What a nightmare: the SHT condition strength was not listed anywhere, since it was never meant to be a final temper. To make matters worse, some of these pins had already been installed [un-plated, not-PHT] in a few acft. Our best guess was the pins were ~1/2--2/3s the required ~185-KSI FTu. Stress weenies sharpened their pencils and accepted these few PIN installations as structurally [barely] adequate; and were judged OK for corrosion sake since these (2) jet were due to be retired in ~15-years [not 30-yeasa as projected for some acft].

ASMH gave me the clue card for an elegantly simple solution: PHT the I-718 PINs to 185-KSI and the diametrical contraction [~0.0008/1.0000”-Dia] would allow a 0.0003—0.0005-thick cadmium [or zinc-nickel] plating build-up [T X 2] to bring-them back to within the required dimensional-range. This solution worked... and was a rare case where 'dumb-luck' saved the day.

Unfortunately the 1.1250”-Dia and 1.2500-“-Dia [220-KSI FTu PHT/Strain-hardened] I-718 bolts that surrounded these shear-PINS were also specified as 'bare' on the installation drawing. These bare-bolts had to be sent back to the OEM for application of cadmium plating, since they had to be re-ground undersize to allow for plating thickness build-up.

FINAL NOTE. The cad-plating and post chromate treatments provided significant isolation between dissimilar materials. However, the issue with serviceability for the next +/-30-years [in a wheel-well environment] forced-us to ‘wet-install’ the PINS [and Bolts/nuts] with permanent chromated sealant… in-lieu-of low-viscosity [temporary] CPC… for the added isolation and corrosion-inhibitors. Then another problem emerged: these PINS were solid, which was ‘OK’ with ‘wet-CPC’ installation! However, the shop guys reminded me that high viscosity ‘wet-sealant’ PIN installation would allow the pins to be inserted only about ½-way before they seized due to ‘hydraulic lock’. With stress’s permission we added a 0.1875-D-hole on the PIN centerline to allow air and sealant to escape when the PINS were pressed-in… which then had to be sealed-off after pin was installed/seated. My confidence level in the long-term corrosion durability for this installation was very high.

Now... gotta-go back to work...

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true.
o For those who believe, no proof is required; for those who cannot believe, no proof is possible.
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion"]
o Learn the rules like a pro, so you can break them like an artist. [Picasso]

 

[tbuelna]

wktaylor- Sorry to go OT, but after reading your "weird" war story about problems with cres shear pins I couldn't resist.

Back in late 1998 I was working on the Shuttle program and the guy in the cubicle next to me was the subsystem manager (SSM) for the orbiter drag chute. STS-95 launched on the evening of Thursday October 29, and right after ignition of the main engines the cres shear pins holding the drag chute protective door in place failed prematurely. The 12lb door fell free, struck the center engine nozzle, and was blasted 1/2 mile away by the exhaust flow. Fortunately the vehicle made it safely into orbit.

When I came into the office early Friday morning I saw every top level Rockwell manager for the Shuttle program huddled in the cubicle next to mine, and they all appeared extremely nervous. I could hear their conversations and it quickly became clear there was a serious problem with the mission. The situation was made especially complicated by the fact that American space hero John Glenn was on board. A bit later when things cooled down I poked my head over the cubicle wall and asked the SSM next door what happened. He explained what had occurred during launch and described the concern there was about the hazard an unsecured drag chute posed during return flight. For example, it would likely be catastrophic if the chute worked free during re-entry and became entangled with the tail. The massive drag chute was compressed into a cylinder housing using a hydraulic press and was deployed by a pyro charge. There was no safe way to jettison the chute while in orbit. But after studying the problem for a couple days it was determined that the best option was to simply leave the chute in place and not deploy it at landing. The official story was that the cres shear pins were manufactured and installed correctly, and they failed for some "unknown" reason.

In the end everything worked out fine for STS-95, but a couple years later the story gets a bit stranger. I was called for jury duty and I was selected as a potential juror for a civil case where the plaintiff was suing his former employer for wrongful termination. The plaintiff was an engineer that had worked on the design of the shuttle drag chute and his former employer was the company that supplied the drag chute. I was no longer working on the shuttle program by then, and during questioning in the jury selection process I made it clear that I had worked on the Shuttle program. Somehow I was still selected to serve on the jury.