|
Joined: Mar 2022
Posts: 7
Member
|
OP
Member
Joined: Mar 2022
Posts: 7 |
Per SAE J443 the “A†strip lower boundary of use is 4A. “N†strips are to be used below 4A which equates to roughly 12N. (3 times) AMS2430 3.11.1 sets the upper limit boundary for drawing specifications that list only one intensity. It states -0 to +30% of the stated requirement or .003 inch whichever is more.
My problem with the AMS 2430 3.11.1 “rule†is that there seems to be an omission with regard to the scaling of the strips. If we were to have a requirement of “peen to intensity 4A†the limits of intensity are 4A to 7A. If one was to move these intensity limits to the N strip scale we would have limits of 12N to 21N. (3 times)
AMS2430 3.11.1 does not account for strip “scaling†therefore if we had a drawing requirement of “peen to intensity 12N†(approximate equivalent to a 4A intensity) our intensity limits get moved from 12N to 21N mentioned earlier to 12N to 16N by applying the 30% rule of AMS2430 3.11.1. This is a reduction of three times in the intensity tolerance zone just by changing from the A strip “scale†to the N strip “scaleâ€. Effectively the allowable compressive stress tolerance band of the peened component is getting penalized three times by changing strip scales from A to N.
This could explain why the industry perceives N strips as extremely sensitive. Should the N strip not be given three times the room to move on N scale? Can you see how dramatically the control would be affected if one was to use an N strip at or in close proximity to a 4A intensity requirement?
SlYdEr
Last edited by Slyder; 03/24/22 02:01 AM. Reason: Typo
Slyder
|
|
|
|
Joined: May 1999
Posts: 200
Member
|
Member
Joined: May 1999
Posts: 200 |
Slyder,
If the engineering drawing specifies and intensity range such as 4A you must use A strips. Conversion to the N scale is not permitted.
N scale strips do provide more resolution for low intensity peening. I've also seen engineering drawings also improperly call out the use of N scale strips to be used at a rather high intensity range of .010-.025N. (.003-.008A) In this example the range on the engineering drawing should have been .004-.008A. But then engineer probably assumed jet engine blades are normally peened on the N scale so that is what was used. But because the wrong scale was chosen the N strip will likely yield more variation since it's being over worked and bending both lengthwise and widthwise and the end result is too much variation.
AMS2430 REV U States: 3.11 Tolerances 3.11.1 Unless otherwise specified, variation from the specified (minimum) peening intensity shall be - 0, +30% to the nearest 0.001 inch (0.025 mm) or 0.003 inch (0.075 mm); whichever is greater. For example, a specified peening intensity of 0.006 inch A denotes an arc height of 0.006 to 0.009 inch (0.15 to 0.23 mm) on the "A" test strip and a specified peening intensity of 0.018 inch N denotes an arc height of 0.018 to 0.023 inch (0.46 to 0.58 mm) on the "N" test strip.
The Unless otherwise specified statement is key. Many engineering drawings have a specific intensity range, example: .008-.012A.
Last edited by Walter; 03/24/22 12:45 PM.
|
|
|
|
Joined: Mar 2022
Posts: 7
Member
|
OP
Member
Joined: Mar 2022
Posts: 7 |
Thank you Walter for your reply.
The real problem I have with AMS2430 3.11.1 is with the “.003 whichever is greater” part of the rule.
If we are to consider why we ultimately need the N strip it is indeed a resolution issue. 4A is the last intensity on the A scale that the Almen gauge has resolution to measure using the 30% rule (for those drawings that state only a minimum Intensity). The 30% rule sets the arc height limits for 4A at .004” to .005”. The difference between the upper and lower limit is .001”. The Amen gauge has a resolution of .0001” therefore using the “rule of tens” from metrology the gauge is capable here but it is the lowest A scale intensity it is good for. So we transition to the N scale which, because it is thinner, bends approximately 3 times as much as an N strip when exposed to the same shot stream conditions. The interesting point to make here is that the Almen gauge is capable to measure the .001” tolerance proposed by the 30% calculation from the minimum 4A. So why grant a tolerance expansion to .003 using the “or .003 whichever is greater” portion of the AMS2430 3.11.1 rule. This makes no sense to me and also unfairly punishes N strip use in these intensity ranges (ranges near the upper limit of the N scale) verses the use of A strips at the lower end of the A scale.
Now let’s look at the 30% rule with regards to N strip measurement. Using the 30% rule of AMS 2430 3.11.1 The Almen gauge again is capable of measuring down to 4N as the arc height limits calculated here are .004” to .005”. Now we have a problem at intensities below 4N. The gauge is no longer capable. Here and only here would would it make sense to offer an “or” to the 30% calculation of AMS2430 3.11.1 and that “or” would not definitely not be “or .003 whichever is greater”. It would be “or .001” whichever is greater”.
In both cases the cutoff for capability of the the Almen gauge for the 30% rule is 4 but perhaps it is not only the Almen gauge resolution that is a factor. Perhaps it is the repeatability of manufacturing the strips themselves. If we to propose a .001 tolerance between the upper and lower limits of either strip perhaps the manufacturing process of the strips themselves would make that tolerance unrealistic/unachievable. Is this the real reason we have this “or .003” whichever is greater” portion of the rule? Is it the strips? .003” seems large to me if this is indeed the case.
Thanks again to anyone who can help me reconcile this portion of AMS2430 in my mind.
SlYdEr
Slyder
|
|
|
|
Joined: May 1999
Posts: 200
Member
|
Member
Joined: May 1999
Posts: 200 |
Almen strips, N, A, and C is not about the capability of the Almen gage it's about the strips themselves an the reproducibility of the process. Think of N strips as ounces, A strips as pounds an C stipes as tons. You need to use the right scale to get the most accurate measurement.
The intensity range is a process measurement, it's what the process is capable of repeating, the range options listed in AMS2430 have nothing to do with strip or gage capability.
"The 30% rule sets the arc height limits for 4A at .004” to .005”
No, The intensity range is .004-.007A True 30% of .004 = 0.0012 and to the nearest .001 would make the range .004-.005A. However, AMS2430 goes on to say that the minimum range is either 30% of the nominal value or .003" whichever is greater. .003 is greater than .001, therefore the range is .004-.007A.
The 30% rule would only come into play if the nominal intensity value was .012 or greater.
.010 would be .010-.013 .110 would be .011-.014 .012 would be .012-.016 This is the first value where 30% would be greater than .003, .0036 actual rounded up to the nearest .001. and so on.
[u][/u]
|
|
|
|
Joined: Mar 2022
Posts: 7
Member
|
OP
Member
Joined: Mar 2022
Posts: 7 |
Hi Walter…. This article in Shotpeener contradicts your first statement about the scale of strips having nothing to do with the capability of the Almen gauge fixture. Is this article incorrect? https://www.shotpeener.com/library/pdf/2013011.pdf
Slyder
|
|
|
|
Joined: Mar 2022
Posts: 7
Member
|
OP
Member
Joined: Mar 2022
Posts: 7 |
As well, in my first post I was thinking of the A, N and C strips as ounces, pounds and tons. I asked why if at 4A we could set the upper limit at 7A due to the “or .003” whichever is greater” could we not set the upper limit of a 12N intensity at 21N? This would be direct scaling but instead when using N strips at 12N we only get to set the upper limit at 16N.
There is something incorrect in the logic. I could get behind the intensity only being able to increase 30% across all scales as ultimately these arc heights are trying to control residual stress characteristics of the components they peen. This “or .003 whichever is greater” does not make sense. Why at the lower end of the A scale would you allow the strip arc height maximum to grow larger than 30%? In fact at 4A the arc height % increase calculates to 75% if .003 is used. On the N scale the application of the “or .003 whichever is greater”, intensity of 1N is allowed to expand to 300% as a %arc height increase.
Slyder
Last edited by Slyder; 03/27/22 09:06 PM. Reason: Typo
Slyder
|
|
|
|
Joined: May 1999
Posts: 200
Member
|
Member
Joined: May 1999
Posts: 200 |
Slyder,
Perhaps it would help me understand the issue better if I knew the answer to this: Are you designing a part and looking to come up with the proper intensity range or are someone who is performing the shot peen processing of a part?
"Why at the lower end of the A scale would you allow the strip arc height maximum to grow larger than 30%? In fact at 4A the arc height % increase calculates to 75% if .003 is used." Reason is process capability. variation in the strips themselves and the process need at least .003 or greater tolerance spread.
The scaling you mention of 3x is theoretical, its not something a shot peen service supplier like myself would consider at all. All I can do is see what the desired intensity range is on the engineering drawing. It might be shown as 4A or it could be shown as something like: .004-.006A, .004-.007A, .004-.008A etc.. It is better when the exact range is on the drawing as it leaves no doubt as to the expectations. Once a peening process is established via the saturation curve and an intensity value is declared, in production that intensity value has to be repeatable within +/-0.0015" of the declared intensity value and be within the specified limits of the engineering drawing. The more locations that need intensity to be both determined and verified the more difficult both become to accomplish. Bottom line is it best if the engineering drawing states the range and not have to rely on the specification default.
|
|
|
|
Joined: Mar 2022
Posts: 7
Member
|
OP
Member
Joined: Mar 2022
Posts: 7 |
Using the ounces and pounds analogy you proposed:
1 lb = 16 ounces
An equivalent 4lb to 7lb tolerance in ounces would be a factor of 16 in ounces so we get 64 ounces to 112 ounces.
Using the AMS2430 3.11.1 UOS rule, the scale we are using (ounces or lbs) does not matter when using the 30% rule. 30% increase is 30% increase so scaling works regardless.
4 lbs x 30% = 1.2 lbs so we get a tolerance using the 30% rule of - 4 lbs to 5.2 lbs (64 ounces to 83.2 ounces)
64 ounces x 30% = 19.2 ounces so we get a tolerance using the 30% rule of - 64 ounces to 83.2 ounces (4 lbs to 5.2 lbs)
These are equivalent tolerance ranges when we use the 30% portion of the rule.
But when AMS2430 3.11.1 says “or .003” whichever is greater” the .003” should specify what scale you are allowing the increase in, A or N, because they are both measured in inches. The .003 should not be applied without this consideration. In our analogy we would have to say “ or 3 lbs whichever is greater” then we could equate 3 lbs to ounces scale as 48 ounces and use 48 ounces to base our decision for the application of the 30% instead:
4 lbs x 30% = 1.2 lbs - 1.2 lbs is less than 3 lbs so our increase defaults to the 3lb addition and becomes 4lbs to 7lbs
64 ounces X 30% = 19.2 ounces - 19.2 ounces is less than 48 ounces (3lbs) so our increase defaults to plus 48 ounces and becomes 64 ounces to 112 ounces.
Because both the A scale and N scale are measured in inches the scaling factor of 3 seems to be getting forgotten in the spec.
Slyder
|
|
|
|
Joined: May 1999
Posts: 200
Member
|
Member
Joined: May 1999
Posts: 200 |
I think you should join https://www.sae.org/servlets/works/ specifically the ASEC Aerospace Surface Enhancement Committee (AMS) and Surface Enhancement Committee (J documents). If you want something changed in the specifications, this is the place to do it. Lucky for you AMS2430 is coming out with a draft revision soon! Possibly in May, you can participate in the meetings which happen twice a year either in person or via Webex. The next meeting is in May 10 Troy, MI United States. You'll see more meeting details if you join. Once the ballot for the draft comes out ,you would post your concern there. What you will need to do is detail exactly how you would like this paragraph to be worded. You must also provide a rational statement detailing why this change is necessary.
|
|
|
|
Joined: Mar 2022
Posts: 7
Member
|
OP
Member
Joined: Mar 2022
Posts: 7 |
Thank you Walter.
To answer you above question: : Are you designing a part and looking to come up with the proper intensity range or are someone who is performing the shot peen processing of a part?
I am the process engineer not the product engineer. I am working with a drawing requirement “peen to intensity of 10N”. There is no upper limit set so AMS2430 3.11.1 becomes very important to me and the cost incurred to maintain the process inside a very tight N tolerance zone relative to the tolerance zone that would be set if my drawing said peen to intensity 4A.. I know… I know… but Forgive me once again but 10N “converts” to 3.3A, an intensity very close to 4A. Our 10N requirement straddles the desired high limit of the N strip and the low limit of the A strip in fact.
Some product Engineers may be relying on this spec to set their upper boundaries for them. I am not saying they are or that this is right or wrong but it may be happening. They may be thinking that the 3N tolerance zone is indicative of peening process capability and is easily maintained at the upper limits of the N scale so they let the spec set it.
My other thought goes back to the manufacturing process of A strips and N strips. We are using the high precision 1S strips. I would think the strip manufacturing tolerances for things like thickness, hardness, modulus of elasticity etc… would be similar or identical in the A 1S strip and the N 1S strip. The problem is that the N strip may magnify the effects of those manufacturing tolerance fluctuations 3 times more than the A strip, thus the reason we see N strips drift significantly from time to time and the industry perceiving them as highly “sensitive”.
Thanks again
Last edited by Slyder; 03/28/22 04:00 AM. Reason: Clarification of point
Slyder
|
|
|
|
|