Logo
Builders' Plan Gallery  |  Hip Pocket Web Site  |  Contact Forum Admin  |  Contact Global Moderator
March 28, 2017, 03:43:04 PM *
Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with email, password and session length
 
Home Help Search Login Register
Pages: [1]   Go Down
Print
Author Topic: Change in rubber energy with temperature  (Read 584 times)
0 Members and 1 Guest are viewing this topic.
PaulR
Bronze Member
***

Kudos: 0
Offline Offline

Australia Australia

Posts: 15

Topic starter


Ignore
« on: March 09, 2017, 08:45:09 PM »

Hi Everyone,

It is well known that the energy of the rubber we use increases with temperature and back in the pre Super Sport days this was found to be a few tenths of a % per degree Centigrade.  

With the current Super Sport rubber the only actual measurements I have seen are my own that were  published in the 2016 NFFS Symposium and reproduced here. Here the actual total rebound energy of the rubber (configured as Coupe motors) was measured over a fairly wide temperature range and showed a very good linear variation of 1.3% per degree C. However, I have often seen a figure of 0.8%/C quoted and was wondering if anyone can point me to some actual measured data. Since this figure is somewhat less than the one I measured I would assume that it relates to a "Figure of Merit" measurement, which from my own measurements I know always gives a lower value than a measurement of the actual total rebound energy of a fully wound motor (and the only one relevant to actual flight performance).

Thanks in anticipation,

Paul.
Change in rubber energy with temperature
Attached files Thumbnail(s):
Change in rubber energy with temperature
« Last Edit: March 09, 2017, 09:31:07 PM by PaulR » Logged
Tapio Linkosalo
Platinum Member
******

Kudos: 25
Offline Offline

Finland Finland

Posts: 1,040



Ignore
« Reply #1 on: March 10, 2017, 01:01:49 PM »

Figure 0.8% per degree comes from my tests of F1B motors with the Pearce stretch test method. I have tested a large number of motors, both Tan II and Tan SS, in ambient temperature conditions, and then fitted a mixed statistical model to the data, using rubber batch and temperature as independent variables. That 0,8% comes from the slope coefficient of the linear regression part of the model (temperature -> energy return). Th figure was estimated separately for TanII and TanSS, and was of same magnitude. I wrote a paper on the rubber tests to NFFS FF sympo a few years ago.

Sorry if y replies will be a bit infrequent in the next few days, just about to start flying for the F1D Euro Champs.

Logged
PaulR
Bronze Member
***

Kudos: 0
Offline Offline

Australia Australia

Posts: 15

Topic starter


Ignore
« Reply #2 on: March 10, 2017, 09:12:02 PM »

Hi Tapio,

Thanks for your quick response and my apologies for not remembering your paper in the 2012 Symposium.

In fact it confirms my suspicion that the 0.8%/C figure relates to a "figure of merit" test and not a full rebound energy test. The reason is, if I understand it correctly, your test pulled the rubber to around 90% of the breaking STRAIN. This corresponds to somewhat less than 50% of the breaking STRESS and so falls into the figure of merit category.  However, my tests impose a stress of around 95% of the breaking STRESS to give a better measure of the total rebound energy. Indeed, to achieve this stress level in a pull test would require a pulling load of well over 200kg!

What does this mean? If the rubber is consistently tested by some figure of merit test, then the appropriate correction for varying temperature is whatever such a test indicates and this will depend upon the maximum stress imposed, in your case the 0.8%/C you measured, though this will be different for tests that use a different maximum stress. However, my figure of 1.3%/C gives a true indication of how the energy of a fully wound motor will vary with temperature. This is quite significant and is no doubt why top fliers like Alex Andruikov heat the motor by breathing into the motor tube prior to final hand turns and launch.

I noted also your interesting measurements of the change in "energy" over time and they seem to reflect the same things I observed: some increase while some others decrease. I have not had any feedback on my conjecture that this might be due to under or over cross-linking of the rubber during processing, and this changing over time as the rubber ages. Do you have any thoughts on the matter?

Regards,

Paul.
« Last Edit: March 10, 2017, 09:39:53 PM by PaulR » Logged
Tmat
Titanium Member
*******

Kudos: 64
Online Online

Canada Canada

Posts: 2,873




Ignore
« Reply #3 on: March 11, 2017, 12:42:23 AM »

Paul,
I have a spreadsheet that Alex Andriukov has used for rubber energy test when using a single small loop of 1/8" rubber. Within the sheet is a correction for temperature given as 0.75% per degree C. This is the value I have used in the past as well although at the moment I cannot recall the exact source. I assume it was from a Fred Pierce article on rubber testing.
If the actual value is as high as 1.3%/C then that is indeed significant.

Tony
Logged

F1B guy...
But don't hold that against me!
PaulR
Bronze Member
***

Kudos: 0
Offline Offline

Australia Australia

Posts: 15

Topic starter


Ignore
« Reply #4 on: March 11, 2017, 12:56:11 AM »

Hi Tony,

Thanks for those comments.

The graph I showed above resulted from full winding tests on Coupe motors, and were not even done sequentially,  i.e. some low temperature tests interspersed with higher temperature ones, just as the weather dictated. The excellent linear correlation gives me confidence that they are quite accurate and repeatable. If Alex didn't apply at least 95% of the fracture stress that would explain the apparent discrepancy, since much of the change lies above the knee. I haven't seen any other full winding tests for comparison.

So, yes I think it is quite significant and if you can blow a few degrees into the tube it will be well worth it!

best Regards,

Paul.
Logged
Tmat
Titanium Member
*******

Kudos: 64
Online Online

Canada Canada

Posts: 2,873




Ignore
« Reply #5 on: March 11, 2017, 01:15:44 AM »

Paul,
Alex's correction factor is just that. A factor to apply to correct for temperature variance during testing. As far as I know it was not determined from the actual tests themselves. He uses a stretch test which applies an extremely high stress. Roughly 13.75 kg for a 1.0 gram sample of rubber strip. A full motor would experience over 170 kg of force!! (Assuming 26 strands of 1/8"). This is of course far higher than the 36 to 40 kg typically used for a full motor force x distance test.

An interesting comment is that I discussed rubber testing with Alex a few years ago and asked if he thought that a full motor test could give sufficient "information" on the rubber energy content compared with his smaller, single loop high stress test. His answer was "no".  Shocked

I read that to mean that a Figure of Merit test using 90% breaking strain would not be sufficient to know the rubber energy content. However, a full motor test using 95% of the breaking stress would likely render the motor unusable afterwards!

Tony

Logged

F1B guy...
But don't hold that against me!
PaulR
Bronze Member
***

Kudos: 0
Offline Offline

Australia Australia

Posts: 15

Topic starter


Ignore
« Reply #6 on: March 11, 2017, 03:57:08 AM »

Hi again Tony,

I fully agree with Alex that you can't get a useful indication of the rebound energy from a full motor that you intend to use afterwards since it is necessary to get near the breaking stress and this will usually damage the motor making it unsuitable further use. That is why I use full (Coupe) motors wound to breaking point to get an indication of the energy of any particular batch. Figure of Merit testing (i.e. testing full F1B motors to loads of only around 30-40kg) can only give a rough relative indication of energy within a batch (though it is not even very sensitive for this purpose) but it is useful for selecting motors to give specific torque and turns.

It is a pity that there haven't been more winding tests up to the breaking stress for confirmation of the actual temperature coefficient under real competition conditions.

Regards,

Paul.
« Last Edit: March 11, 2017, 05:02:24 AM by PaulR » Logged
PaulR
Bronze Member
***

Kudos: 0
Offline Offline

Australia Australia

Posts: 15

Topic starter


Ignore
« Reply #7 on: March 11, 2017, 08:35:49 AM »

Hi Again Tony,

Prompted by your comments I have looked up the spreadsheet that Alex sent me a couple of years ago (maybe the same as yours?) and it indicates that the actual load applied to the loop was 7.96 kg, if I am reading it correctly. Through some scaling that I don't fully understand this becomes 13.75 kg, but that is not the actual value. If 7.96 kg is the actual load, it represents about 80% of the fracture stress since a loop of 1/8" usually breaks at a load just under 10kg. This is certainly high enough to give an indication of the total rebound energy but still a bit short of the 95% I used to get the temperature coefficient. As you say, he quotes a value of 0.75%/C but there is no indication of where this came from. It would be nice to know if he actually measured this or was simply quoting from some other source.

Also, just a correction to an earlier comment, of course the breaking load of a full 26 strand x 1/8" F1B motor would be around 130kg, not the 200kg I mentioned. Blame it on "Old-Timers"!!

Regards,

Paul.
Logged
duration
Gold Member
*****

Kudos: 6
Online Online

United States United States

Posts: 351



Ignore
« Reply #8 on: March 11, 2017, 09:28:07 AM »

Paul,

I've been wading through all my files on rubber testing and found a few things that might be of interest, if not of help:

* A published chart by Fred Pearce from the 1980s states "Energy storage varies almost linearly with temperature. For example FAI rubber measuring 3400 foot pounds per pound at 75 degrees F would measure 3070 at 50 degreesF and 3730 at 100 degrees F."  (When I visited Fred in the late 1990s, his test rig was outside, under a roof overhang. Houston, Texas, can be well over 100 degrees F in the Summer. I have no idea if Fred tested motors at various ambient outside temperatures and adjusted readings or if tested when the outside temperature was closer to the 75 degree F standard.)

* An article by Mike Evatt in October 2010 Aeromodeller mentions a 0.5% power loss or gain for a 10 degree F change in temperature. He does point out the advantage of stretching the motor out and immediately winding to take advantage of the warming due the stretching. In hot weather the motor out, then relaxed to cool the rubber slightly before winding.

* One thing from Evatt's article surely must be a typo: "These temperature swings can be as much as 250F for a full wind-up and and about a third of that for a hard stretch." I can't imagine a 250F increase in rubber temperature during winding; 25.0F seems more likely.

* R.W.New's paper "The Physical Behavior of Rubber Strip Motors to F1B Specifications, & The Effect of Heat Treatment on Performance" from a 1980s Sympo (I think it is from a Sympo, but all I have is a photocopy) suggest three interesting ways to increase energy yield; one involves heating the the rubber in castor oil to just below the boil for 20 to 30 minutes, a second methods is to wrap a castor oil-soaked motor in aluminum boil and bake in the oven at 70 degrees C for 45 to 60 minutes, the third method uses brown soft soap dissolved in water and simmering at 60 degrees C for 20 minutes. (If anyone want's to stink up their kitchen, I can email a copy of the article.)

Louis
Logged
PaulR
Bronze Member
***

Kudos: 0
Offline Offline

Australia Australia

Posts: 15

Topic starter


Ignore
« Reply #9 on: March 11, 2017, 07:00:13 PM »

Hi Louis and others,

"Cooking" the rubber sounds a bit drastic and if it worked would imply that in the '80's the rubber was not vulcanized (cross-linked) to the optimum degree. I don't think I would try it with the current Super Sport.

I just realized that Alex's scaling is to normalize all his tests to a standard of 1g of rubber 100mm long (i.i. different cross section to the 1/8" norm.)  That is why the maximum load in the normalized data is higher than actually applied.

Paul.
Logged
gil43
Copper Member
**

Kudos: 0
Offline Offline

Israel Israel

Posts: 8



Ignore
« Reply #10 on: March 16, 2017, 11:06:50 AM »

For general knowledge (and for my spreadsheet calculations..)
what is the basic temperature? (is it 25 deg C ?)

Nice to read about Deg C and not Fahrenheit  Roll Eyes
Logged
Hepcat
Platinum Member
******

Kudos: 131
Offline Offline

United Kingdom United Kingdom

Posts: 1,519



Ignore
« Reply #11 on: March 16, 2017, 02:08:32 PM »

Quote from #2: "top flyers like Alex Andruikov heat the motor by breathing into the motor tube prior to final hand turns and launch"

I was surprised to see this as the FAI rules say no heat shall be applied to the motor.

John

Logged
DerekMc
Platinum Member
******

Kudos: 49
Offline Offline

United States United States

Posts: 1,162



Ignore
« Reply #12 on: March 16, 2017, 03:09:33 PM »

Quote from #2: "top flyers like Alex Andruikov heat the motor by breathing into the motor tube prior to final hand turns and launch"

I was surprised to see this as the FAI rules say no heat shall be applied to the motor.

John



I'm not sure how effective the breathing into the tube is for warming up the tube itself. I timed for one of Oleg Kulakovsky's  fly off flights during the Fab Feb at Lost Hills earlier this year. He blew a into the motor tube a couple times before loading the wound motor. Not enough to raise the temp a large amount I would think. Fliers often put loaded half tubes in their coats to warm up the rubber before winding.

The rule about heat being applied to the motor was a response to the practice (at the time) of using battery powered heaters on the motor tube. That did raise the temperature quite a bit. That was before half tubes and winding outside the plane became common.
Logged

They fly better when you smile!
Derek
PaulR
Bronze Member
***

Kudos: 0
Offline Offline

Australia Australia

Posts: 15

Topic starter


Ignore
« Reply #13 on: March 16, 2017, 09:32:26 PM »

I have observed a number of interesting practices on colder days, including holding the motors near the body as Derek noted, leaving them in a warm car, leaving the loaded but unwound half tube on the stooge in the sun and covered to shelter from any wind, breathing into the motor tube, and so on. Just how effective these are would depend upon how rapidly the motor reaches ambient temp when being wound and loaded, given the the additional complication of the heat given out when rubber is stretched and absorbed again when it is allowed to relax (maybe this can be optimized with winding technique, as mentioned by Loius?). Black fuselages would also be better than light colored ones! Probably not too important for the rounds but could be the winning edge for a cold morning fly-off?

Paul.
Logged
Pages: [1]   Go Up
Print
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Valid XHTML 1.0! Valid CSS!