BRIT NERF Forum Index
Join! (free) SearchFAQMemberlistUsergroupsLog in
Falcon 130 Motors: 3S LiFe and 4S LiPo

 
Reply to topic    BRIT NERF Forum Index » Modifications -> Chrono Data Forum View previous topic
View next topic
Falcon 130 Motors: 3S LiFe and 4S LiPo
Author Message
Boff
The Dealer


Joined: 08 Dec 2012
Posts: 3392
Fav. Blaster: (Modified) Stampede ECS
Location: Bristol, UK



Post Falcon 130 Motors: 3S LiFe and 4S LiPo  Reply with quote
Usual work in progress, I've only done preliminary tests and will continually update this as I add new information.

--

Previous work on the Falcon 130 motors have yielded successful results in determining optimum battery configurations and performance. The motor platform itself is being more and more widely adopted with no known failures at the time of writing. In terms of muzzle velocity, 3S Lithium Polymer (LiPo) has been previously determined to be superior to 2S LiPo. However, further exploration of battery options is warranted and recent discussion around Falcons raised the possibility of 3S Lithium Ion Phosphate (LiFe) batteries and 4S LiPo. The aim of this post, therefore, is to explore these options and compare them to extant data sets.

Lithium Iron Phosphate battery packs have a nominal voltage of 3.3V per cell compared to Lithium Polymer's 3.7V per cell. It is expected that a 3S LiFe pack with a nominal voltage of 9.9V will perform slightly under its 3S equivalent due to a lower RPM as a result of lower voltage. However, previous experience with LiFe battery packs (albeit largely ancedotal) has seen fewer failures with LiFe chemistries compared to the higher voltage LiPo packs. Additionally, LiFe chemistries are more stable than LiPo and have a lower self-discharge capacity which lends them to being a little more user friendly for the less experienced player.

It's important to note that these trials focus on muzzle velocity as the sole indicator of performance. It is understood that this is not the sole indicator of performance and that the human will indeed be at the centre of the Nerf hobby, irrespective of the hardware they have access to.


Method:

Falcon 130 Data Collection:
Data collection for the Falcon 130 motors themselves was done using our Competition Electronics Pro-chrono Digital shooting chronograph with custom lighting rig, placed on a tripod out of direct sunlight with no other artificial light sources present. Shots were fired over the chronograph at a rate of 2/s (unless otherwise specified in the specific protocols).

Constants:
Darts: Koosh (eBay seller Yutoys, sourced March 2015)
4S LiPo: ZIPPY Compact 1500mAh 4S 25C Lipo Pack
3S LiFe: ZIPPY Compact 1800mAh 3S 30C LiFe Pack

Stryfe Configuration
Batch number: 42484

The Stryfe was rewired using a Blastersmiths UK solid state wiring kit and switches. In the place of a micro-switch, a IRF44ZN MOSFET was used to control the flywheels using the stock harness as per the previous method. The same batteries were used for all trials so the pack was held outside the blaster using Velcro on this occasion.

Comparative Data Sourcing:
Data for comparisons was sourced from previous work I've done on Koosh shoots and from OldNoob's fantastic Modified Blaster Data Archive. The data sources for each can be found below.

Comparative data for these trials were sourced from the previous trials linked here.

Preliminary trials:
Each battery configuration was subjected a preliminary test before the main trial was conducted. Each configuration was tested with an 18 dart magazine filled with Koosh darts, fired at the standard 2 darts/s. After each preliminary test, the motors were checked for signs of damage and smells of ozone. Neither pack configuration to be tested showed any signs of damage and both configurations were cleared for further testing. A 10 minute gap was left between trialling the two configurations on the test Stryfe.

It was also determined during the 4S LiPo test that ear protection was necessary to prevent problems with frankly banshee like screaming that occurs when you put 4S through a Falcon 130. All 4S tests were carried out with cotton wool in ears.

Results:

All Falcon 130 data can be found here in one convenient document.

Stryfe 3S LiFe Results:
Stryfe 3S LiFe Trial #1:
Data is presented in the first column.

Average: 115.42FPS

Stryfe 3S LiFe Trial #2:
Data is presented in the second column.

Average: 117.16FPS

Stryfe 4S LiPo Results:
Stryfe 4S LiPo Trial #1:
Data is presented in the first column.

Average: 105.58FPS

This trial was particularly frustrating to perform. Darts routinely failed to read on the chronograph with lots of fish-tailing and other inconsistencies. In all, 74 pack-fresh darts were fired in order to get the 50 results detailed above. The rest were errors or not read at all by the chronograph. It is also worth noting that the chronograph got a much greater dusting of foam than the 3S LiPo or 3S LiFe trials.

Stryfe 4S LiPo Trial #2:
Data is presented in the second column.

Average: 105.68FPS

This trial was done around an hour after the previous one to stop me from wanting to strangle a kitten in frustration. Similar difficulties in Trial #2 as with Trial #1 - 62 darts fired in total this time out. It's important to stress that the LiFe and 3S LiPo trials had no failures (hence not recorded) so this is something significant. Additionally, Trial #2 recorded a similar covering of foam over the chronograph as Trial #1 and higher than I recall it being for the 3S LiPo trials.

Analysis:
Falcon 130 3S LiPo vs Falcon 130 4S LiPo:

Aggregate data of both sets of trials was used in this analysis to give the largest possible sample size.

3S vs 4S LiPo - #1 and #2 Trails:
Average 3S LiPo FPS: 112.86FPS
Average 4S LiPo FPS: 105.63FPS
Student's t-test (p < 0.05 is statistically significant) : 0.0002052661799

Falcon 130 3S LiPo vs Falcon 130 3S LiFe:

3S vs 4S LiPo - #1 and #2 Trails:
Average 3S LiPo FPS: 112.86FPS
Average 3S LiFe FPS: 116.29FPS
Student's t-test (p < 0.05 is statistically significant) : 0.03870631442

Once again, an aggregate of both data sets was used for analysis.

Discussion:

3S LiPo vs 4S LiPo:
The 3S Lipo vs 4S LiPo trials yielded a set of resutls that saw a reduction in muzzle velocity for the 4S pack. In addition, a greater number of mis-fires and general dart problems were seen in both trials with the 4S pack. While the 3S trials proceeded without incident, 24 and 12 extra shots were required respectively to attain a full data set for each trial. These mis-records are at odds with the experience in the 3S LiPo and LiFe trials where regular, repeatable shots were possible and no mis-reads occurred. This factor alone is sufficient to suggest a preference for 3S LiPo over 4S LiPo. It is suggested the drop in muzzle velocity is down to a slipping effect on the flywheels. That is to say the flywheels are spinning too fast for efficient force transfer inside the flywheel cage. Further tests are needed to fully explore this 'flywheel slip' phenomenon. Again, there is a reasonable amount of anecdotal evidence to suggest this as a mechanism for sub-par performance at higher rotational velocities but a full characterisation is yet to be done. Most critically, these data suggest a 'sweet spot' for flywheel rotational velocity and possibly torque that will challenge the notion of 'more is better'. It is already known that certain builds, including certain 180 builds, will result in dart damage due to flywheel interaction wearing down the foam and so optimisation of blaster configurations for each given motor and battery configuration is still warranted. Until further exploration and characterisation can be attained, there are no cut and fit rules in this area.

3S LiPo vs 3S LiFe:
Indeed, such a lack of cut and fit rules are illustrated by this somewhat unexpected result. The 3S LiPo data set for the Stryfe averages at 112.86FPS while the 3S LiFe set averages at 116.29FPS. These data challenge the original hypothesis that a higher RPM would lead to a higher muzzle velocity. Again, it is not so clear cut. While the 2S LiPo and 3S Lipo clearly scaled linearly, there is clear evidence presented here that suggests 3S LiFe packs may offer a higher muzzle velocity, albeit not by a great deal. Both packs used have similar constant discharge capacities (55A for LiPo and 54A for LiFe) so current is most likely available in excess (citation probably needed).

On testing protocol:
The protocol adopted in this set of trials was an improved one over the previous Rapidstrike and Stryfe trials. This iterative approach to protocols has led to a much cleaner and larger data set. While the previous trials, in particular the case of the Rapidstrike, had several data sets that were collected but did not meet the standard for analysis, these trials have been much more consistent. It's important to note that the introduction of preliminary trials to determine to hone the exact protocol of the main trials is a key stone in this improvement. Provided these preliminary tests are monitored and their outcome recorded in the main write up, they are a useful tool in optimising trials both in terms of quality of data collection and in reducing cost of darts for each trial.

Future prospects:
The Falcon 130 motor is steadily proving itself as a solid motor for use in blasters. Its popularity is boosted by the fact there is no need for shell modification. Further work, as noted above, into the 'flywheel slippage' phenomenon is required but trials to explore this might difficult to construct. Slightly more accessible will be the required work on LiFe battery packs to determine if their capacity of higher specific power has applications beyond simple response times. Additionally, it is worth exploring how these packs might be used to optimise blasters on the field. The limiting factor in their adoption is that LiPo packs as they are known are much more diverse in size and shape compared to the relatively narrow range of LiFe packs. Of course, all this work is only about honing the 10% of the Player equation that relates to equipment, actual field experience and team work will overcome a great number of faults in the real world and that is something we should always keep in mind.


Disclosure:
The author is an employee of Blastersmiths UK Ltd and owns shares in the company.


A note on data ethics and publication bias:
I'm committed to the principle of All Trials which essentially means that all the tests I do get published; not just the successful ones or the comparative ones. If I've done it it'll be here on BritNerf. The more data we have, the more useful things we can determine by analysing it. Something I might miss, someone else might notice because of the data I've published. Yes, I'll use the successful data when I'm promoting a mod or upgrade but the bad stuff will be here, too.

_________________
Boff: Managing Director, Blastersmiths UK & BUZAN Founder (formerly)
| Blog: nerfarmourer.tumblr.com | Website: www.blastersmiths.co.uk | Legal: http://pastebin.com/6sQ7c3jg |

Last edited by Boff on Thu Nov 19, 2015 4:54 pm; edited 13 times in total
Thu May 21, 2015 11:56 am View user's profile Send private message Visit poster's website
OldNoob
Foam Data Collector


Joined: 07 May 2013
Posts: 4833
Fav. Blaster: Xtreme 180 Rapidstrike
Location: In the Boonies



Post Reply with quote
Thanks for keeping the tests coming. More data is always useful.
I noticed a few overly high shots (clear mis reads) had crept in. I usually delete the highest and lowest shot from my data set before writing up, to cut out derps or misreads. If you are using the indoor rig I made, be very careful of interfearance from exterior light sources or low sun from your window. I would remove the 150's and 130's from the data as they are clearly an error. That much deviation is also going to bend the averages.

_________________
Big_Poppa_Nerf wrote:

Boff whats the damage? I have spent over 3 times my Nerf budget this month already. Part of me is trying to be a responsible parent/husband/house owner. The other half is just says 'Ahhhh, Screw it!'.
Thu May 21, 2015 12:09 pm View user's profile Send private message Send e-mail
Boff
The Dealer


Joined: 08 Dec 2012
Posts: 3392
Fav. Blaster: (Modified) Stampede ECS
Location: Bristol, UK



Post Reply with quote
I publish my data in its entirety and work from there. If people want to rework it post hoc then that's fine with me. I usually note the outliers in my discussion section. The joy of these much larger data sets is that the outliers will have less of an effect than if we were working with a data set of 10 or 20. Smile

_________________
Boff: Managing Director, Blastersmiths UK & BUZAN Founder (formerly)
| Blog: nerfarmourer.tumblr.com | Website: www.blastersmiths.co.uk | Legal: http://pastebin.com/6sQ7c3jg |
Thu May 21, 2015 12:18 pm View user's profile Send private message Visit poster's website
OldNoob
Foam Data Collector


Joined: 07 May 2013
Posts: 4833
Fav. Blaster: Xtreme 180 Rapidstrike
Location: In the Boonies



Post Reply with quote
Fairy snuff.
I will bear all that in mind when I create the comparison data for all the common motors once my house roof is done. We still have better data raw than every other forum.

_________________
Big_Poppa_Nerf wrote:

Boff whats the damage? I have spent over 3 times my Nerf budget this month already. Part of me is trying to be a responsible parent/husband/house owner. The other half is just says 'Ahhhh, Screw it!'.
Thu May 21, 2015 1:03 pm View user's profile Send private message Send e-mail
TopDart25
Junior Member


Joined: 24 Mar 2015
Posts: 40
Fav. Blaster: Rapidstrike / Stryfe
Location: United States



Post Reply with quote
Thank you for this data. Now I am seriously considering the Falcon 130 motors.

Do you think that a build up of foam on the flywheels from previous tests could have contributed to the higher average fps on the 3S LiFePO4 battery vs the 3S LiPo?

Clearly, the 4S LiPo performed worse than either of the above as you mentioned darts could be slipping on the flywheels. I am just wondering if maybe 3S LiFe or 3S LiPo are in that "sweet spot" range that you mentioned or if 3S LiPo is too much and darts are also slipping at that speed contributing to the lower average fps.
Fri May 22, 2015 12:18 am View user's profile Send private message
OldNoob
Foam Data Collector


Joined: 07 May 2013
Posts: 4833
Fav. Blaster: Xtreme 180 Rapidstrike
Location: In the Boonies



Post Reply with quote
If I remember correctly the layer of foam on both blasters was already pretty well established, it stays constant as any surplus is fed through the system as foam dust! I have had both of the test blasters through my mitts at some point for paint.
Over a shoot sample of this size, with nearly everything else constant, whichever shoots the hardest is the sweet spot. If it's an awkward pack for fitting in trays then you may have to accept the less perfect pack to have one that fits.
What you want with the flywheels is them to spin at the best "transfer speed" where as much of the energy as possible is transmitted to the dart with minimal loss of flywheel speed. Often this does not correspond to the absolute screaming motor top end. I guess this is what you can see between 3s and 4s.

_________________
Big_Poppa_Nerf wrote:

Boff whats the damage? I have spent over 3 times my Nerf budget this month already. Part of me is trying to be a responsible parent/husband/house owner. The other half is just says 'Ahhhh, Screw it!'.
Fri May 22, 2015 6:56 am View user's profile Send private message Send e-mail
Boff
The Dealer


Joined: 08 Dec 2012
Posts: 3392
Fav. Blaster: (Modified) Stampede ECS
Location: Bristol, UK



Post Reply with quote
TopDart25 wrote:
Do you think that a build up of foam on the flywheels from previous tests could have contributed to the higher average fps on the 3S LiFePO4 battery vs the 3S LiPo?


Unlikely. These flywheels were primed with 400 shots before the first round of trials (the 2S versus 3S LiPo) which consisted of 200 shots. Then they were taken to GC1 where I fired well over 500 rounds (it was a light weekend, all things considered) and then brought back to the office, derped around with a little bit and then run through the preliminary and main tests. By my count that's well over 1000 shots before the start of the LiFe trial so if we're adding extra material after that number of shots I'd be very surprised.

OldNoob is right, sourcing a correctly rated and sized pack for the Stryfe might be a pain in the plums. I do, however, have the option of considering this for my Rapidstrike. I might see if the data are replicable across to other platforms in due course.

_________________
Boff: Managing Director, Blastersmiths UK & BUZAN Founder (formerly)
| Blog: nerfarmourer.tumblr.com | Website: www.blastersmiths.co.uk | Legal: http://pastebin.com/6sQ7c3jg |
Fri May 22, 2015 7:52 am View user's profile Send private message Visit poster's website
OldNoob
Foam Data Collector


Joined: 07 May 2013
Posts: 4833
Fav. Blaster: Xtreme 180 Rapidstrike
Location: In the Boonies



Post Reply with quote
The other option would be a small PWM in the Stryfe, if you can get one small enough to fit with the 3s pack in the tray expander. Then you can tune your flywheel max speed.

_________________
Big_Poppa_Nerf wrote:

Boff whats the damage? I have spent over 3 times my Nerf budget this month already. Part of me is trying to be a responsible parent/husband/house owner. The other half is just says 'Ahhhh, Screw it!'.
Fri May 22, 2015 8:29 am View user's profile Send private message Send e-mail
Boff
The Dealer


Joined: 08 Dec 2012
Posts: 3392
Fav. Blaster: (Modified) Stampede ECS
Location: Bristol, UK



Post Reply with quote
That'd be fine on paper but in practice, I'd be a nightmare. Also, the whole reason I'm on the Stryfe at the moment is in an effort to keep things simple.

It might be theoretically possible to have your blaster tune your optimum flywheel speed with an inbuilt chronograph and funky stuff like that. No, we're not exploring that avenue, I'm just being silly. Razz

_________________
Boff: Managing Director, Blastersmiths UK & BUZAN Founder (formerly)
| Blog: nerfarmourer.tumblr.com | Website: www.blastersmiths.co.uk | Legal: http://pastebin.com/6sQ7c3jg |
Fri May 22, 2015 8:49 am View user's profile Send private message Visit poster's website
SSGT
Nerf Addict


Joined: 12 Jan 2013
Posts: 1273
Fav. Blaster: 'Rave-N' CS-18, Elite Alpha Trooper (EAT) CS-12
Location: Durham



Post Reply with quote
That shouldn't be too hard to implement actually. A light gate in the barrel and a FET controlling the motor via a microcontroller. Razz

One of the main problems it might have is that maximum velocity isn't always the optimum velocity. It might measure 115fps in the barrel but, depending on the type of dart, it might fishtail and flop out of the sky 10ft away. There's not really any way of solving that unless you tell the mcu an average velocity to aim for and, if it can't achieve it, just aim to optimise it after that.

You could go one further and adjust the RoF of the RS to balance RoF and muzzle velocity. Not exactly practical but almost worth doing just for the fun of it. Depends on your definition of "fun" I guess. Razz

_________________
Trust me, I'm an engineer! Very Happy

Antonius wrote:
Compoooter bits make shiny Nerf blaster go whoosh!
Fri May 22, 2015 9:24 am View user's profile Send private message
Antonius
Regular Member


Joined: 24 Sep 2013
Posts: 228
Fav. Blaster: Modded Nite Finder EX-3, Stryfe
Location: Cork, Ireland



Post Reply with quote
As an added bonus, with all the additional cruft added to the front end of the barrel, you can always use the blaster as a cudgel!

Laughing  Razz

_________________
Two Wrongs don't make a Right…
Three Lefts do though…
Fri May 22, 2015 1:19 pm View user's profile Send private message
Boff
The Dealer


Joined: 08 Dec 2012
Posts: 3392
Fav. Blaster: (Modified) Stampede ECS
Location: Bristol, UK



Post Reply with quote
SSGT, I know it's not that difficult; I just didn't want anyone asking us to build them one.

On topic, I've fleshed out the discussion a little more. There's still more to come, me thinks. I'm pondering how we might construct a trial to calculate optimum rotational velocities for a given motor or torque value. I actually think using the light gate system with barrel velocity as a proxy for muzzle velocity and field behaviour might be the best way to go on that. You could have the micro-controller pair the duty cycle of the motor (proxy for RPM) with the barrel velocity quite easily and graph it accordingly. If you were then to slave an RS pusher box to a FET, too, you'd be able to get consistently timed shots. With an O-scope, you could even go for full electrical characterisation, too. That'd be a lot of time and data, however - not something we can do in the near future. Unless anyone has a scope floating around they'd be willing to lend me?

And this is the fun part of my job. I get to design, build and test things according to market need. The market needs 130 motors and that's what we're doing here. The boring bit comes when I have to fill out tax returns and deal with the lawyers. Very Happy



_________________
Boff: Managing Director, Blastersmiths UK & BUZAN Founder (formerly)
| Blog: nerfarmourer.tumblr.com | Website: www.blastersmiths.co.uk | Legal: http://pastebin.com/6sQ7c3jg |
Sat May 23, 2015 10:34 am View user's profile Send private message Visit poster's website
Display posts from previous:    
Reply to topic    BRIT NERF Forum Index » Modifications -> Chrono Data Forum All times are GMT
Page 1 of 1

 
Jump to: 
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum

Card File  Gallery  Forum Archive
Powered by phpBB © 2001, 2005 phpBB Group
Design by Freestyle XL / Flowers Online.
Create your own free forum | Buy a domain to use with your forum