PDA

View Full Version : New site attempts to test upgrade parts



AustinWolv
10-20-2004, 09:30 AM
A new site has shown up that is apparently going to help people understand upgrade part combinations, as well as directly compare respective parts to see how they hold up.
It will be interesting to see how they conduct their testing since there are multiple variables involved........

http://www.airsoftmechanics.com/

Obsidian
10-20-2004, 10:09 AM
thats going to be an expensive site to run.

AustinWolv
10-20-2004, 10:42 AM
thats going to be an expensive site to run.
Agreed. I'm keeping a living chart of all the parts I go through (rounds lasted, failure mode), and it isn't exactly a short-term task.

In their most recent update, they already ate a sector gear. Moreover, they are going for the extreme case, as a M150 spring is being used. Problem with that is they are now testing the piston for strength really. For example, most players run M120 springs. Thus, the fault in the testing that I could envision follows as this:
1. Two pistons, one with metal teeth and one with only one metal tooth
2. M150 spring testing shows that the one with metal teeth lasts longer, whereas the one with only one metal tooth gets the plastic teeth eaten up

That likely doesn't correlate to the following testing:
1. Same as #1 above except use a M120 spring
2. Impact strength of piston with only one metal tooth and the rest of the teeth being plastic is sufficient enough to take a M120 spring just as well as the piston with all the metal teeth.

One could argue that the piston with metal teeth may last longer, but it also could be argued that premature wear will occur on the sector gear because of that.

What I'm getting at is that the testing is looking for upper-spectrum tensile and shear strength, but is skipping over impact strength to do so. This may be a reasonable assumption on their part however.

JayZ28
10-22-2004, 02:12 AM
Hello Everyone-

I am the crazy person daring to attempt the AirsoftMechanics site. Yes it has already been quite an expensive site to run, but being the first to try something often is. The goal of course is to achieve a certain degree of credibility with my tests, hence attracting readership and subsequently sponsorship. Now on to the concerns.

I apologize for the sparse details on testing procedures currently posted on my site, but rest assured I am doing more than throwing each piston into a gun and holding the trigger down for a while to see which breaks first. I am also testing material hardness with a Barber-Colman Impressor, which is a reasonable measure of material tensile/yield strength. Thus far, the "softest" piston has been the Classic Army yellow piston, and not surprisingly it has also failed most spectacularly. The effect of piston weight has also been taken into account in the measured rate of fire and muzzle velocity, which should both be reduced given heavier pistons. Thus far the results have shown this correlation to be somewhat true.

The reason for conducting these tests under an extreme case was two-fold. Firstly, I wanted a stressful enough condition to ensure failure of some of the pistons in the given timeframe so that meaningful comparisons can be drawn. I didn't think anyone would want to read a review where every piston performed equally, which I contend would be likely given an M120 spring, with the possible exception of the Classic Army piston. Secondly, it is very likely that a piston capable of withstanding an extreme case can also withstand a less extreme case. While the best solution would be to test every piston with every spring, time and money constraints serve as formidable barriers, even for a diehard like myself.

AustinWolv makes a very valid point in suggesting that pistons with full metal teeth will likely display a strong bias in my tests as they are unlikely to fail, and this could only be countered by noteing that full metal teeth will inevitably lead to increased sector gear wear, hence reducing their desirebility. Wear describes a surface interaction, and while rate of wear is not easily quantifiable directly in this case, it is in general highly correlated with the inverse of the relative hardnesses between the two surfaces in question, favoring plastic toothed pistons when sector gear wear is a concern.

Aside from rate of wear, the plastic toothed pistons will still display a range of differering degrees of failure, and those that fail "less" under the extreme case will arguably be even more capable in a less extreme case, and thus score higher on the imaginary scale of desireability.

Lastly, with regard to "impact strength", I assume you are referring to the ability of a piston to resist cracking during impact. While I lack the equipment to directly measure brittleness, the correlation between material hardness and brittleness is generally accepted, in that harder materials tend to be more brittle than softer materials. As I understand it, brittleness describes what happens after the yield strength of a material has been surpassed. Materials that are more brittle will fail catastrophically (crack) when their yield strength is surpassed while less brittle materials will fail gradually (deform). I contend that in either case, the piston in question has failed. Whether the piston cracks and causes immediate failure, or it deforms gradually to the point of malfunction at a later time, the piston has failed. Since harder materials also exhibit a higher yield strength, I contend that they will "last longer" in that they are less likely to fail compared to a softer piston given the same forces. This of course is a partial assumption, and hopefully my test results will make these assumptions clearer.

Thank you all for your interest in my humble site. My goal is to serve the airsoft public, and to that end I impore you all to not hesitate with suggestions, questions, concerns, corrections or anything else you think might be helpful. AustinWolv seeing as you are a mechanical engineer, I would be happy to hear any thoughts and suggestions you might have. I'm an engineer myself, but of the electronic variety.

Jay
www.AirsoftMechanics.com

AustinWolv
10-23-2004, 01:25 PM
JayZ28, what I'm referring to in regards to impact strength is the ability of a material to withstand shock loading, as well as resistance to fracture (cracking) during shock loading (an impact event). The true test of that would be an Izod impact test, whereas the Impressor measures hardness, which is more related to wear. Impact strength is worthwhile IMO for our pistons since the sector gear is whacking the hell out of them at every pick-up rotation.

Hardness is a reasonable measure in this case, as yield strength can be loosely derived from hardness numbers, and it is of course resists surface degradation between the two components.

However, I contend that toughness is more important for our application, since that is the ability of a material to resist permanent distortion and to withstand shock events (absorb energy).

The piston is clearly a component that needs to resist shock. It is first subjected to an impact with the first tooth of the sector gear whacking the hell out of the back tooth of the piston. That is a common failure in the semi-circle piece of piston that includes that back tooth is ripped off, where the rest of the piston teeth still look in good shape. At this point, the other teeth engage and the piston is in a tensile state with the spring pushing it forward, but the sector gear pulling it backwards. Then, you have the piston slamming forward into a hard stop situation in hitting the cylinder head.
Thus, you have a component that is subjected to impact, shock, tensile forces, and compressive forces. Hence, it makes me think a piston being tough is more important than it being strong. Why?

The toughness allows the material to deform some in order to dissipate energy. This is valuable in that your material can take more energy (shock, for example) before failing than a component that is strong. For example, I have had plastic pistons hold up better under M140 springs than aluminum pistons, because the plastic teeth will deform slightly allowing the sector gear to pass by more easily, whereas the metal teeth, while obviously much stronger, would not deform causing a catastrophic failure.

All of that above is ideal. Obviously, you can't have the best of both worlds in getting both a strong and tough material. There are tradeoffs. Thus, that is why your testing is a good venture in that it will allow buyers to weigh what they really need.


AustinWolv makes a very valid point in suggesting that pistons with full metal teeth will likely display a strong bias in my tests as they are unlikely to fail, and this could only be countered by noteing that full metal teeth will inevitably lead to increased sector gear wear, hence reducing their desirebility. Wear describes a surface interaction, and while rate of wear is not easily quantifiable directly in this case, it is in general highly correlated with the inverse of the relative hardnesses between the two surfaces in question, favoring plastic toothed pistons when sector gear wear is a concern. Absolutely and a good summary, as I think you'll find that most people would rather kill pistons than gearsets.
It will also be noted that metal teeth, while stronger, are not as tough. Hence, while they'll take a lot of force for stronger spring situations, they'll also fail catastrophically.
This takes me back to the point that a plastic piston might be a better option for those running lower-powered springs because that material dissipates the energy better which is energy that isn't imparted into the mechbox shell or other components for example.

Your assumptions look solid, so it will be of interest to see the results.

The other aspect I'm curious about is how are running controlled piston tests? In other words, how are you eliminating other variables? For example, if you are running Piston A with gearset A and Spring A, and then gearset A fails for whatever reason, I would assume you are putting the same type of gearset in that mechbox (call it gearset A.1) to finish running Piston A. In addition, I'm assuming you are running that piston to failure? Are inspections being conducted at certain usage intervels?
The other aspect I'm curious about is this:
Let's say you are running polycarb Piston A at a M150 spring, and it lasts 1000 rounds.
Then, you run a reinforced polycarb Piston B at a M150 spring, and it lasts 10000 rounds.
The assumption is that Piston B is better, ie. stronger. However, going back to the toughness and impact strength comments above, how do you quantify that Piston A and Piston B perform at the same level with a M120 spring?

Again, I understand your assumptions and limitations on testing resources, but just pointing out one aspect that is being overlooked.

Just wait until you get to gearset testing. Ouch. ;)

Like I mentioned above, I'm keeping a living spreadsheet of all the parts in my guns to see raw MTTF, measured in rounds fired. It is not completely scientific of course in that the parts are in different mechboxes with different components, but I eliminated variables by charting the individual parts. The goal isn't to say which part is definitively better than another, but to give an idea of part life in general.

Anyway, kudos on your venture, as I definitely deem it valuable as another resource for people to consider. Let us know of any help we can provide.

JayZ28
10-27-2004, 02:20 AM
Thanks for the reply AustinWolv. This is precisely the type of
discussion I was looking for from a mechE. Feel free to set me
straight if I'm making any bad assumptions.



However, I contend that toughness is more important for our application,
since that is the ability of a material to resist permanent distortion
and to withstand shock events (absorb energy).

I am not entirely clear on this here, as it is my understanding that
"toughness" refers to the area under the stress-strain curve for a
particular material. The ability of a material to resist permanent
distortion (staying within the elastic region of the curve), while
obviously related to toughness, seems to me to be more described by the
yield strength, since the elastic region ends at the point of maximum
yield strength and permanent plastic deformation occurs afterwards. Toughness to me is more a measure of the total abuse a material can withstand prior to catastrophic failure. While a material that is tougher can
take more abuse by dissipating the energy through permanent deformation, my argument is that even though the failure isn't immediately apparant, it
has already begun and will continue to progress until failure. Of course the rate of progression could be slow enough to make the piston "last longer", or it could be fast as in the case of the Classic Army piston. The key goal of my tests is that hopefully results will show a range of failure, and readers will be able to decide which piston suits them best.



The toughness allows the material to deform some in order to dissipate
energy. This is valuable in that your material can take more energy
(shock, for example) before failing than a component that is strong.
For example, I have had plastic pistons hold up better under M140
springs than aluminum pistons, because the plastic teeth will deform
slightly allowing the sector gear to pass by more easily, whereas the
metal teeth, while obviously much stronger, would not deform causing a
catastrophic failure.


In your reference to toughness allowing the material to "deform some in order to dissipate energy", and in particular the reference to plastic teeth deforming to allow easier passage of the sector gear teeth, I assume you refer to plastic deformation? This brings up a very interesting situation which I had not previously considered.. the ability of a tougher piston to automatically "correct" for slightly mis-aligned/incompatible piston-gear combinations through plastic deformation to a more compatible configuration, relieveing the otherwise increased stress caused by the mis-alignment, and hence avoiding catastrophic failure.

Your experiences with the M140 spring is precisely the situation I'd like my results to answer. Obviously, if the M140 spring is able to crack your aluminum piston, the stress it is placing upon the piston exceeds the yield strength of that particular aluminum alloy. Given that aluminum has a higher yield strength than polycarbonate, one can safely assume that the stress placed on the piston is also higher than the yield strength of polycarbonate. This indicates that any deformation which occurs to either piston must be plastic. Since the aluminum piston failed catastrophically but the polycarbonate piston did not, the amount of elogation caused by the M140 spring must've been greater than the fracture point of aluminum, but less than the fracture point of polycarbonate. This is certainly a possibility since polycarbonate is tougher than aluminum, but the interesting question becomes, just how much deformation will the polycarb piston take before it fails completely?



This takes me back to the point that a plastic piston might be a better
option for those running lower-powered springs because that material
dissipates the energy better which is energy that isn't imparted into
the mechbox shell or other components for example.

If mechbox shell longevity is a concern, there is no doubt that the
softest piston will be the most beneficial, since it will absorb the
most energy. But of course, no one is suggesting the use of Classic
Army pistons precisely because they are just too soft. The bottom line
is that it's a tradeoff, and hopefully the results of my tests will help
in revealing just the right combination of attributes which gives the reader a
balance of the factors of concern.



The other aspect I'm curious about is how are running controlled piston
tests? In other words, how are you eliminating other variables? For
example, if you are running Piston A with gearset A and Spring A, and
then gearset A fails for whatever reason, I would assume you are putting
the same type of gearset in that mechbox (call it gearset A.1) to finish
running Piston A. In addition, I'm assuming you are running that piston
to failure? Are inspections being conducted at certain usage intervels?


Each piston is installed in the same mechbox with the same components, tested for rate of fire and muzzle velocity, then subjected to the full auto stress test. In the full auto test, the pistons are cycled for 1000 rounds, at 100 round intervals with a cool down time in between each interval. Each piston is then removed for examination and the results recorded as the "first pass" results. After each piston has been tested, they will then each be reinstalled for additional 1000 round passes. So far, after the first pass, most pistons displayed limited damage, with the noteable exception of the Classic Army yellow piston. I believe 2nd pass and 3rd pass results will start to differentiate the pistons more definitively. I have not considered running the pistons to failure, although that is definitely an option.

Yes, the same gearset A.1 is the same as gearset A (Systema All-Helical Super Torque-Up).




The other aspect I'm curious about is this:
Let's say you are running polycarb Piston A at a M150 spring, and it
lasts 1000 rounds.
Then, you run a reinforced polycarb Piston B at a M150 spring, and it
lasts 10000 rounds.
The assumption is that Piston B is better, ie. stronger. However, going
back to the toughness and impact strength comments above, how do you
quantify that Piston A and Piston B perform at the same level
with a M120 spring?


Obviously it's impossible to say definitively that piston A and piston B will perform equally under a different set of conditions, but, even given your toughness and impact strength comments, clearly piston B, having been able to withstand 10x more impacts than piston A without failure on an M150, would easily withstand whatever impact an M120 spring could impart. If the pistons are similarly priced (as most of the pistons in this test are), it would undoubtedly be a better investment to purchase piston B.



Just wait until you get to gearset testing. Ouch. ;)


I'm already there :) Purchasing every Prometheus and Systema helical gearset is quite expensive, in case anyone hasn't figured that out for themselves.



Anyway, kudos on your venture, as I definitely deem it valuable as
another resource for people to consider. Let us know of any help we can
provide.

Thanks again for your interest. Definitely keep me honest if I'm making any assumptions that are incorrect in your view. My limited material sciences knowledge certainly makes me susceptible to bad assumptions. It's great to have a fellow engineer double check my work.

-Jay

JayZ28
10-27-2004, 11:01 AM
After checking out "Mechanics of materials: an introduction to the mechanics of elastic and plastoc defomration of solids and structural components by Hearn, E. J.", I realized that there is another critical element missing: fatigue. Given the cyclic nature of the piston in the AEG mechbox, even if the subjected stress is well below yield strength, fatigue still can occur from localized stress concentrations, initiating crack formation. Fatigue strength is difficult to quanitfy since it is based on many factors including the physical design of the piston and minor surface imperfections, but it does have strong correlation with toughness. Tougher materials are more resistant to fatigue under cyclic loading, and this certainly favors the use of polycarbonate pistons in instances where the loading stress is below yield strength. As you stated earlier, since my tests represent an extreme case, strength is emphasized over toughness in the short term, since the pistons must first withstand the increased tensile forces to even survive the test, thereby masking the effects of fatigue strength. Currently, most of the polycarbonate pistons are holding up quite well and the aluminum pistons have showed no damage, but it is unknown to me whether I will be able to cycle either piston type enough to surpass their fatigue limits.

-Jay

AustinWolv
10-27-2004, 01:11 PM
Ahh, too much reading! :)

as it is my understanding that
"toughness" refers to the area under the stress-strain curve for a
particular material. That it does. Hence, the toughness of a material is dependent on both the elastic and plastic regions of the stress-strain curve if I recall correctly. Been a while since I've dealt with materials study, but a brittle material's ultimate strength will occur so suddenly (no elongation/plastic deformation) that it won't be as tough since the area under the curve is limited as compared to a material that withstands greater strain %.

my argument is that even though the failure isn't immediately apparant, it
has already begun and will continue to progress until failure. Correct, with some materials reaching that point (ultimate strength) well before others. That comes down to material choice, even amongst plastics, which the Classic Army piston may be an example of:
http://www.dow.com/styron/design/guide/mechanical.htm
See how the straight polycarb is elongating over a much longer period before it fractures compared to the PC/ABS sample?

This should be of interest in terms of strength versus toughness: http://www-materials.eng.cam.ac.uk/mpsite/physics/introduction/default.html
As you can see, while polymers will not approach aluminum in strength as expected, they can be tougher, which for high-cycle applications is beneficial.


in particular the reference to plastic teeth deforming to allow easier passage of the sector gear teeth, I assume you refer to plastic deformation? I think that may play a part in it, along with general surface wear.

This brings up a very interesting situation which I had not previously considered.. the ability of a tougher piston to automatically "correct" for slightly mis-aligned/incompatible piston-gear combinations through plastic deformation to a more compatible configuration, relieveing the otherwise increased stress caused by the mis-alignment, and hence avoiding catastrophic failure. Bingo. That is what I was getting at earlier in the thread, but not very concisely or clearly. In contrast, that aluminum piston tooth will not deform as easily, thus you'll see teeth just get sheared off if enough force is applied.


Your experiences with the M140 spring is precisely the situation I'd like my results to answer. Obviously, if the M140 spring is able to crack your aluminum piston, the stress it is placing upon the piston exceeds the yield strength of that particular aluminum alloy. Given that aluminum has a higher yield strength than polycarbonate, one can safely assume that the stress placed on the piston is also higher than the yield strength of polycarbonate. This indicates that any deformation which occurs to either piston must be plastic. Since the aluminum piston failed catastrophically but the polycarbonate piston did not, the amount of elogation caused by the M140 spring must've been greater than the fracture point of aluminum, but less than the fracture point of polycarbonate. This is certainly a possibility since polycarbonate is tougher than aluminum, but the interesting question becomes, just how much deformation will the polycarb piston take before it fails completely?
Just to clarify, the teeth being sheared off of the pistons in both cases were the failure modes, just as most people experience with piston failures (along with the back semi-circle section that is commonly ripped off failure mode). To me, that makes sense in that the piston will withstand much greater longitudinal stress than shear stress.
I think you'll find upon your inspections that there will be "wear grooves" in the plastic pistons where they've meshed, but you won't find the depth of those grooves in the aluminum pistons/metal toothed-pistons. Over time, the gear-mesh groove in the plastic teeth will continue to get deeper until the sector gear teeth will bust through and strip the plastic teeth off.

But of course, no one is suggesting the use of Classic
Army pistons precisely because they are just too soft LOL! I wouldn't suggest the use of any CA internal parts. :D

As far as test procedure, sounds like you got it nailed. If parts are replaced, do you apply fresh lubrication to the new parts or depend on the residual lubrication from the existing mating part?

Progressive reliability testing with periodic checks and limiting variables is good methodology.

What number of samples are you running per part? Again, I understand the cost limitations involved, but one-off testing of a particular piston isn't statistically "kosher". This is premature, but have you considered possibly doing reliability confidence limit testing in the future, so you could avoid having to run significant sample sizes?

Fatigue is a fairly detailed topic because there are so many variables: surface finish, temperature, material, type of load.....
The thing about fatigue is that it isn't even necessarily the load cycling alone, but design, manufacturing defects, and/or process changes can obviously contribute.

My limited material sciences knowledge certainly makes me susceptible to bad assumptions. I doubt my material background is up to snuff either. :o

JayZ28
10-29-2004, 01:19 PM
As far as test procedure, sounds like you got it nailed. If parts are replaced, do you apply fresh lubrication to the new parts or depend on the residual lubrication from the existing mating part?

I always apply fresh lube for each new piston, and this has actually led to some problems with too much lube getting into the cylinder and out the nozzle, reducing some muzzle velocity readings. I figured it's worth it to give each piston the best chance it has to succeed.



What number of samples are you running per part? Again, I understand the cost limitations involved, but one-off testing of a particular piston isn't statistically "kosher". This is premature, but have you considered possibly doing reliability confidence limit testing in the future, so you could avoid having to run significant sample sizes?

That's a great point, but unfortunately, sample size can't be addressed in the near future since I'm running on a limited budget. The money that goes into this is basicaly reducing me to eating lots instant noodles for dinner :) If the site takes off however and I have enough sponsorship, I definitely wouldn't mind increasing the sample size. The lack of sample size will definitely be noted in the review.

-Jay

AustinWolv
10-29-2004, 01:27 PM
The money that goes into this is basicaly reducing me to eating lots instant noodles for dinner Maybe you can pull a Jared and get the noodle company to sponsor you as a weight loss marketing ploy like Subway. ;)

id0l
11-02-2004, 06:33 AM
Ahh, too much reading! :)

Damn, you can say that again! I got to this exact part, and decided to read the rest tomorrow. I have learned alot already. I look forward to your site Jay, as I am - like alot of people - appreciative of factual testing and scientific results (even if I don't understand all the super-technical operations)...it sure isn't the "this piston is better because it is aluminum!!!" discussion all over again.

-Sorry to intrude on your discussion-

:)

JayZ28
11-02-2004, 04:05 PM
Thanks id0l. A new replacement gearset is in so the testing has begun again. I should have results ready by next week.

-Jay

JayZ28
12-08-2004, 01:19 PM
I have some updates and more results up on my site if anyone is interested.

-Jay

id0l
12-09-2004, 02:49 PM
Thanks for the update, that was the most in-depth article(s) I have read on pistons. It makes me feel a lot more comfortable with the Deepfire Full TiN piston I received by mistake (as opposed to the HurricanE vented full aluminum piston). I wonder how these will perform with Prometheus EG gears in regards to sector gear/teeth wear? Guess I will have to find out, although I will be running a PDI 170% spring so it won't be as drastic as a m140. Can't wait for your gear set review. :D