RC-Monster Forums

RC-Monster Forums (https://www.rc-monster.com/forum/index.php)
-   Brushless (https://www.rc-monster.com/forum/forumdisplay.php?f=12)
-   -   Kyosho Inferno GT2 VE - Big-Block Edition (https://www.rc-monster.com/forum/showthread.php?t=32204)

Dr_T 06.19.2015 11:21 AM

Kyosho Inferno GT2 VE - Big-Block Edition
Hi all, I'd like to share my ongoing Kyosho GT2 build here. Already burnt up the Turnigy XK 5682 - 900kv motor within running 3 batteries, so currently looking for a good replacement to fit the big-block motor mount. Hope you like it, recommendations are very welcome. I am not aiming for any speed-run records, I'm just trying to build a hassle free ~100 mph set-up that can be ran for an entire pack without having the electronics melting down.

  • Motor: vacant... ;
  • "Ruggedized" Castle XL2 ESC + 4x 680 uF caps + 60x60 mm 24V fan;
  • Turnigy 5.0Ah 6S and possibly 7S 60C (HD series) batteries;
  • Gearing: tbd... Turinigy was geared 49/36 (pinion/spur) for no-load speed of ~162 km/h (6S) with measured 972 kv (with Castle's Torque Control motor test) at 10 timing;
  • Turnigy 1258TG servo;
  • Turnigy 7.5A HV UBEC;
  • 500k Traxxas WT front diff; 40k cSt rear diff; center spool on KYO IG-108 2-Speed Shaft;
  • 800 cSt shock fluid, stock RTR springs;
  • 2 rear toe brace (KYO IF-124B);
  • GRP GMY01-S5 tires, 2014 compound (figured car would be heavy enough to use S5 instead of S3);
  • Parma X-citer #1222 buggy body, hoping it will reduce the typical blow-over tendency of the full GT bodies.

The Looks:

If Batman had a GT2 :)


Under the Hood:


Gearing a 900kv motor on 6S required some creativity... More info on that here: http://www.rc-monster.com/forum/showthread.php?t=32150

Traxxas Revo spur on IG108 2-Speed Shaft (use with use the IG101 100mm Center/Front Driveshaft)

First attempt at some DIY parts

Shortened Mamba XL2 with stronger heatsink clamping and bigger fan. CF bottom plate to help support the part that pushes the foamie on the brainboard.

All mounted up

Fans are 24V, wired through balance lead extension, so I can switch easily between 6S and 7S. This way also prevents the internal BEC of the XL2 from putting power on the fans when using the Castle link. Something that seems to cause potential issues: http://www.offroad-cult.org/Board/lo...40.html#289268. The UBEC is connected to the ESC battery leads, but it has a power switch, so it can be turned off when using the Castle link. With hindsight I should have just gotten a HV servo, and power it from the balance leads as well, in order to avoid the whole need of an external BEC.

Dr_T 06.19.2015 11:27 AM

The Testing

Wet road test drive, skip to 3:24 for stupid-gearing/low-speed Mamba XL2 cogging test.

Weird thing is that the logs show peak motor RPM to be ~25k, which would translate to roughly 1100 kv @ ~100-120 A load. In the zoomed part in the third pic below, you can see the motor RPM going from 0-25k in 0.8 s; with current gearing, that 25k motor RPM corresponds to ~188 km/h wheel-speed... I know I spun the wheels a lot on the wet road, but that seems impossible. In the fourth pic below it seems like it almost takes a second for the ESC to lock commutation after throttle going through neutral. Then with smooth throttle build-up, according to the ESC, the motor revs up gradually from 0 to ~24600 RPM (with this gearing 185 km/h, 115 mph wheel-speed) in 3.3 seconds at 21.5 V, pulling 110 A. 24600 RPM / 21.5 V =1144 kv. Maybe the motor was on its way out already?





Dr_T 06.19.2015 11:37 AM

Cremation of the XK 5682 motor

I guess when Hobbyking is blowing off stuff for peanuts, there's probably a very good reason for it. In all fairness though, the motor was only rated for 106A, so maybe I just had it coming. Log partial with dry road testing shows ~50 A average, ~190 A peak. I thought I could get away with it with some good cooling, but the winding resistance appears just to be too high, causing them to heat up too quickly. On a positive note, it means I'll be looking for a nice replacement motor. RPM in log was unexpectedly high again.




Also tried the new GRP slicks, and as Lizard already mentioned in his 1717 Slash thread, they're less grippy than the older treaded versions... The rubber already feels less sticky and it seems they wear down differently too, the slicks appear to stay smoother (pic below is both S5 compound, used on same surface, same conditions). Contacted GRP about it and here's the reply:


in January when we have put on the market the new line 2015 of 1:8 Gt we have make same little change to all the compound for works this better and also have more time life.

So the compound of 2015 season it is little differnt from the 2014 tyres. But the REVO 2015 and SLICK 2015 compound are the same.
A bit bummed GRP doesn't mention differences in compounds in their specs, but stuff happens.


Dr_T 06.19.2015 11:43 AM

Search for new motor

Maybe I should have looked into it a bit earlier, but after burning up the Turnigy, I got my crayons out and made some graphs to get a better understanding of how much Power is actually required to obtain a certain speed, within a certain time, or distance; thought I’d share it here for later reference and for anyone interested. Basically there are 3 types of Forces at play when accelerating to a certain speed:
  • Acceleration Force, proportional to weight and acceleration;
  • Aerodynamic Drag Force, proportional to air density, frontal surface area, and quadratically dependent on speed;
  • Friction and rolling resistance Forces, proportional to speed.

Power = Force * Velocity, so from these Forces, we can get an idea of the Power requirements. The first graph below shows the Force and Power components as function of speed for a 5 kg RC, with 30cmx15cm frontal area, accelerating at 10 m/s^2 (based on Lizard’s 1717 Slash: 0-150km/h in 4 seconds) through standard airmass. Drag coefficient, Cd, is guessed to be 0.35, same as the default in BrianG’s online calculator. Combined friction and rolling resistance coefficients are the biggest ‘fudge factor’ in the model. I chose them to have the Drag and friction/resistive Forces cross-over at ~50 km/h - just a shot from the hip, no idea how valid that is. The model is just a very simple and straightforward approximation, based on constant acceleration to top-speed (been way too long since I bothered with differential equations :D). In reality, the maximum traction Force the tires can put on the road is limited obviously, so the net longitudinal Force that is available for acceleration, will be limited and, in addition, will reduce with increasing speed due to the growing part of the velocity dependent components.

Dashed lines in the graphs are the individual components (acceleration: a, Drag: D, friction and rolling resistance: f&r), solid cyan lines are the totals; solid magenta lines are totals minus acceleration, so representing the constant velocity Force and Power requirements. You can see how much the acceleration part actually contributes to the total Power requirements, and this model does not even take into account the (acceleration of) rotational masses (just needed a quick fix, might look at that later… well, probably not :D). Accelerating slower would lower Power demands, but running-space would become an issue then quickly. Even with 10 m/s^2 (about 1g), it takes about 100 m to accelerate from 0-160 km/h (100 mph). 0-200 km/h would take ~155 m… so accelerating slower would leave very little time for driving at top-speed (and getting a good reading of it out of a flimsy 1 Hz GPS logger to brag around with on the interwebz :D).


Next step is taking into account system efficiency, also a bit of a ‘fudge factor’. After some tweaking and comparison with results of BrianG’s calculator model, I put efficiency at 69%. Main difference with the scriptasylum model is that that model seems to levy the friction and rolling resistance into the "overall efficiency" (from battery input Power to usable output Power). If I put the friction and rolling resistance coefficients in my model on zero and take an overall efficiency of 63.7%, it seems to produce the same results as the online tool with electrical efficiency set to 85% and 4WD drive configuration - at least for the couple of speeds I checked, in the range that I’m interested in.

The graph below shows the electrical Power and Current (@6S) requirements as function of speed. Based on these approximations, at 6S, it would take about 217 A to get the 5 kg example car (which happens to closely resemble my GT2 of course ;)) with 10 m/^2 to 160 km/h; sustaining that speed would only take 72 A. Continuing up to 200 km/h would take 315 A , plus/minus ~30 A for a kilo more or less and 135 A in steady state. Seems that beyond 160 km/h, things are becoming a bit more tricky.


Long story short: I guess it takes at least a 220-250A motor to be able to do 160 km/h (100 mph) comfortably and reliably with a low-profile 5 kg car on 6S, which also explains the premature death of my 106A Turnigy... good thing it was only 10 bucks.

Feedback and motor recommendations are appreciated. Bear in mind the approximations above are based on a lot of simpliciations and assumptions, I have yet to check it with some logs I have of my 8ight-T to see whether it all makes sense :).

lincpimp 06.19.2015 02:19 PM

Well, first off I would say you need to go up to 8s with the XL2 esc, or higher if you can find a controller to do it... Higher voltage means less amp draw for the same load.

Next, you want a motor that will turn at least 40k rpm, and preferably higher, like 60k. So you will need to go to a very well made (expensive) motor such as a Neu or whatever else is out there that is made well and balanced (4 poles design). A better motor will be more efficient and the higher rpm will allow for shallower gearing, and that will lower your amp load throughout the speed range. Your 900kv motor died due to overgearing... If you had put 12s through it may have survived (half the amp load) and your pinion could have been half the size thus startup load would be greatly reduced. You could probably do this with a Neu 1527 with kv around 1800kv on 8s will give around 50k rpm max and should have enough torque and mass to endure what you are throwing at it....

It would be nice to know what the horsepower (wattage) requirement would be to get to speed, and then to maintain speed. A that point it would be easy to determine the voltage needed with the amp capability of the battery and motor and esc that are available.

But I know that higher voltage and more motor rpm range, and thus lower gearing is the way to go.

bdebde 06.19.2015 02:55 PM

Neu ftw

Dr_T 06.19.2015 04:41 PM

Thanks for the feedback guys! Neu, or Lehner of course (go Europe! :)), sounds nice, maybe one day,... after selling one of my kidneys :). 6S use here was because the motor was only rated 6S (due to excellent rotor quality I quess). Also I'm a bit paranoid of frying my XL2 and want to give it some head-room, also under heavy breaking Voltage bumps (pdf alert). I do have a 7S pack already though, just no >6S charger yet. Might only be a marginal gain in Current reduction, but better than nothing.

You guys really think the kv is the problem? In the end what counts is the Power a motor is able to deliver right? Kv just determines over how wide of an RPM range that Power is spread out. I can't imagine a big fat Castle 2028-800kv, spinning below 30k RPM max, would not be able to pull even taller pinion/spur combos than what I used?

I think main factor the motor died was just the poor quality. If you look at a snip out of one of the logs posted above, you can see that the load with that gearing, even at start-up, actually isn't all that high: 60 A average, 160 A max... should be peanuts for a proper motor, no?


I already have a 1527-1500kv size motor (TP4070 - "poor-man's Neu 1527") in my truggy. It's geared for ~91 mph no-load nominal and it needs very careful throttling and rolling starts to stay under 200 A (6S); it's pulling over 220 A frequently with that gearing (logs below), but it holds up well. I'm really impressed with the performance of that motor, but I more-or-less built the GT2 around the big-block motor mount, so it needs a 56 mm diameter motor :). TP56xx are likely candidates, but not sure yet about kv and size (weight). Would have been nice if Castle had a motor between 1717 and 2028 size.




Lizard 06.20.2015 02:23 PM

Now that you had bad luck going with the low KV/RPM route maybe really try the opposite? The newer TPPowers seem to have higher RPM ratings, even the 5860 can go up to 50k. I'm wondering if those reinforced magnet holders the MBP-RC guy is talking about on his page are now being used again for all TP motors or only the MBP-RC versions. Maybe get one from him if you want to be on the safe side with high RPMs.

Oh, and there is a guy on a german forum who got a TP4070 2200KV and is running it on 6s with an XL2 in a Robitronic Hurricane. Seems to hold up so far but not too much info there about gearing, temps etc.

Dr_T 06.21.2015 03:17 PM

Hmmm..., so here I was thinking I had it all figured out and wanting to prove the world my jumbo pinion was not a stupid idea... LOL... Maybe it is indeed better to go a bit higher kv...

A motor that spins twice as fast, needs to deliver only half of the Torque in order to put out the same mechanical Power (Torque*angular velocity of shaft). Torque is proportional to I/kv, in other words, Current is proportional to T*kv; so with half the Torque and double the kv, the Current actually stays the same. The difference however is that the the higher kv motor, has less windings / turns and thus lower copper resistance (which adds up more when it's a quality motor)... So from a copper (I^2*R) losses point of view, RPM should indeed be as high as possible. Non-copper (core and mechanical) losses however, increase with increasing RPM, so the big question is: where's the sweet-spot?

Bruno actually has a 2200 kv TP5670 on sale, that needs a 400+ A ESC for full potential... so maybe not :). MBP has what seems to be an old (yellow 3 piece design) TP5840S-4Y, 1600kv (31 V max, 270 A, 715 g). There's also a TP5660-6D 1420 kv (old specs: 35V max, 285 A, 835 g) I'm considering. 36/36 gearing will probably be a bit too tight, but 36/40 and 40/40 light and cheap Revo gears will fit and give nice speed range:


Have to sleep on it :).

Dr_T 06.21.2015 03:53 PM

What seems weird looking at all the specs is that D-winds seem to not only have higher Rm (D winds need sqrt(3) times more turns to get same kv as Y winds), but also higher Io (indicating higher magnetic losses), both indicating less efficiency, no?

Skrew it... just ordered the TP5660-6D 1420 kv (old specs: 50k RPM rotor, instead of 53k); 160 bucks shipped seemed like an ok deal at APS; the one at MPB would be 30 bucks more and from TP EU 40 bucks more (US still has low prices, but shipping and customs would ruin that). Now let's just hope I get the right one :D.

Edit 2:
Thought I'd just post this first myself... for the record, the 56 millimeters are not ribbed :).

Lizard 06.22.2015 08:33 AM

Great, can't wait to see how it goes. I really hope you get what you ordered, according to TPPowerEurope, the 6D is 1260kv.

But why no cooling fins version? You wanted to have something that is in stock?

ruudxd 06.22.2015 09:26 AM

Sorry if it is a stupid question, but if you have a bigger pinion then spur, aren't you going away from the 1:1 ratio again?

Dr_T 06.22.2015 02:01 PM


Originally Posted by Lizard (Post 429583)
Great, can't wait to see how it goes. I really hope you get what you ordered, according to TPPowerEurope, the 6D is 1260kv.

But why no cooling fins version? You wanted to have something that is in stock?

No fins is because I cheaped out: 5660 equivalent with fins (5840L, apparently shorter rotor, but same weight and Power, so more copper?) is over 200 EUR currently. TP's changing their specs all the time, the one I ordered seems to be a previous version; maybe that's why it was cheaper. Too bad the table doesn't list Rm:


Bit of a gamble, but we'll see. It's no custom work, so if I get a different motor, I'll send it back.


Originally Posted by ruudxd (Post 429584)
Sorry if it is a stupid question, but if you have a bigger pinion then spur, aren't you going away from the 1:1 ratio again?

Not a stupid question at all Ruud! Having a bigger pinion than spur just makes the gearing taller than 1:1, so with the center drive-shaft spinning even faster than the motor shaft. There's nothing 'magic' about 1:1 gearing, it just happens to work well with the typical 40 mm, high kv, motor set-up used by speedrunners. If motor RPM goes down though, you need taller gearing to achieve the same wheel-speed. Kershaw actually has some over the counter pinions up to 60T for use on lower kv 1/5th scale motors: http://kershawdesigns.com/KyoshoGT-2-5882Mount.htm.

Lizard 06.22.2015 02:22 PM

Hmm, you mean the last number is rotor length and the finned ones have shorter rotors? Now that would add another layer of confusion :)

Always thought they were the same because in the data tables they are listed both in the model no. column.

I think same weight is just wrong, that would be a too unlikely coincidence that two motors with different cans and endbell designs etc. have exactly the same weight.

Dr_T 06.22.2015 02:34 PM

Yeah, maybe their motor designations just don't make any sense...

All times are GMT -4. The time now is 07:11 PM.

Powered by vBulletin® Version 3.8.1
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.