Kyosho Inferno GT2 VE - Big-Block Edition

[ruudxd]

Ah, did not think of it that way, that sounds good indeed.

I hope the new motor serves you well!
May I ask wich Eagle tree you use?
And where did you buy it?
And am I correct to see that you have a mill?
If yes, May I ask wich one?
Because I want one myself soo bad :)


Keep up the good work!

[Dr_T]

Thanks!

Got the EagleTree at a lucky eBay auction, was sold by a shop in Italy. It's the 150A version; I added 8AWG shunts to improve Current rating and lower Resistance (more info):



It's not a real mill unfortunately. It's simple Lux/Wolfcraft drill stander with 2-axis coordinate table (no idea how to call those things :), see pic below). It can be used for very simple milling though, I used it to mill down the battery tray I bought to better fit my needs. Turns out that POM material is quite easy to work on; might try and make a new tray for my 8ight-T by myself. If you really want to make your own CNC mill, there's a thread on URC by Bill Delong about that.

[ruudxd]

Thanks for the info man!
Is it the V3 version?

Maybe CNC later, I like the hand crafted work mote somehow.
Maybe to show to the older generation that there are kids out there that love to get dirty hands, and make nice things :)

[Dr_T]

No, v4 even; not sure, but I think v4 has bigger memory for longer logging. Been wanting to get that nice 10 Hz GPS sensor for it, which should be big improvement over my 1 Hz bicycle logger, but too many other projects needing money more urgently :).

[ruudxd]

Ah, that was a good deal then indeed!
Keep up the good work, thanks for all the info!

[Dr_T]

Received the motor. Seems to be the correct one, even has new 2 piece can, so size-wise it's same as current TP5660s: 80 mm long, instead of the 92 mm of the old specs. Measured kv with the XL2 to be ~1500, so even slightly hotter than spec. Will do some more tests to get a bit of a feeling for timing vs no-load Current later.

Pics below show how much more copper the TP has over the Turnigy motor. Despite having same size and lighter (vented) can, the TP weighs ~70 g (9%) more. There's also a comparison pic of the motor wires, the cutted piece on the bullet connector is from the Turnigy, looks kinda poor compared to the TP's; I actually had to dremel-out the holes on my 6 mm bullets a bit to get the TP wires to fit. But, in all fairness, the TPs cost about twice as much as the Turnigy would have at regular price. Hope the TP can live up to the expectations raised by specs and appearance :).

Still intrigued by the similarity of the appearance and specs of the Turnigy XK5682 and Leopard 5682. Would be interesting to see if anyone has some nude pics of the 56xx/58xx Leo motors.

[Dr_T]

Did some simple no-load testing of the TP5660; used the XL2's datalogging only, which appears to have a bit too low of a Current resolution for accurate figures, but I was too lazy to solder TRX-to-XT150 adapters for my Power meter and you probably get the picture.

Seems those D-winds are pretty power-hungry compared to Y-winds indeed, with quite high no-load Currents. Compared to my TP4070 with identical kv, but Y-wind, I measured this TP5660 D-wind to have about 2.5 times higher no-load Current. Measurements seem to correlate more-or-less with the new TP5660 specs though (although I'm not sure what the reference Voltage is in those specs).

First plot shows max throttle response @ 14.1 V, for 0, 2, 4, 6, 8, 10, 12, and 14 deg timing. Second plot shows some DIY step response with a 6S pack at storage Voltage, for 0 (left) and 5 deg (right) timing. Last pic shows numeric summary of the datalog plots, plus a rough plot of the relationship between the no-load Power and RPM. Interestingly, it seems to take more Power (I*V) to spin the motor with a certain RPM at 14.1 V than at 22.2 V. As I translated the differences in Io already to Power, they should be fully comparable, so can't make any sense of that yet... maybe the ESC is involved in these differences too?

[ruudxd]

Seeing your interest in wires, recistance, winds etc I am thinking you should find a long outrunner and rewind that yourself?

The difference seems to be huge indeed, but I think that is what you pay for to, the material?
Copper is expensive ( at least here in The Netherlands it is )

[florianz]

Great pics thanks!! What a massive motor, my god. I can confirm that y- an d- wind motors behave kind of different. And what I have read and what I have experienced is that they react differently regarding timing. My TP motor d-wind works best with 7,5 degree. I think aveox, Castle motors have only y-wind for a long time now (dunno if that's true).
With what timing are you running the TP's?

[Dr_T]

Originally Posted by florianz

Great pics thanks!! What a massive motor, my god. I can confirm that y- an d- wind motors behave kind of different. And what I have read and what I have experienced is that they react differently regarding timing. My TP motor d-wind works best with 7,5 degree. I think aveox, Castle motors have only y-wind for a long time now (dunno if that's true).
With what timing are you running the TP's?

Yeah, not sure what I'll be running timing-wise yet. Probably very low 0-5 deg. I have been running my TP4070 (supposedly Y-wind) on 7.5 deg on a HW Xerun, but never tested thoroughly if that could be optimized a bit more. Would be nice if more ESCs had dynamic timing, where timing ramps up as RPM increases; been playing around with that on the VESC a bit.

[Dr_T]

Alright, time to get some life in this again. Stumbled on the Basher BSR Rally / Team C GT8LE-RA's 42T spur gears on Hobbyking, and thought they might work as a nice cheap and lightweight pinion/spur combo.

Having seen Lizard's disappointing results with the TP5840 and the resulting philosophizing about how a theoretical max efficient ERPM would translate into lower preferred RPMs for 6-poles (aka, I might have chosen the wrong kv again :D), I got rid of the my cool-looking, but not so smart DIY CF motor spacer, so the motor-mount is mounted directly onto chassis for better heat transfer. Gap between motor and chassis is filled with thermal pad, which also nicely helps cushioning vibrations. As this will only be used on-road, I won't seal up the venting holes as I did with my TP4070.

Still need to add some o-rings to the center-dogbone dirve-cups, as there's a bit too much vibration in those still, but getting close to driving again.

[Dr_T]

Had a little test-drive today: motor power has evolved a bit faster than my driving skill and it was a very big switch to get used to after having mainly played low-speed off-road lately :D.

Ran 42/42 gearing for a no-load speed of 185 kph @ 3.7V/cell on 6S, based on measured 1510 kv. Wanted to run 2 6S packs and 1 7S pack, but had to stop halfway the second 6S pack, because the gears (both pinion and spur) came loose and got chewed-up a bit. Had some throttle glitches in the beginning, with sudden brake pulses while accelerating, which might have been a contributing factor to that, but I think all-in-all the tiny M3x6 screws holding the gears to the spur mounts are just too small. Have to see how I fix that. On the spur-side, I can add lock-nuts, but on the pinion-side there's not any space for that... to be continued...

Anyway, the important part: managed to get it up to 160 kph; motor wasn't breaking a sweat: max temp I saw was 48°C @ 17°C ambient, but I was changing settings a lot and not driving continuously. When everything is dialed in and with some more space and practice, I will be able to push it harder and temps will obviously rise, but from what I can tell so far, I think this motor won't become the bottleneck and I will be killing my batteries before the motor. Looking at the log data, I think I never really reached steady-state speed (need a bit more space and guts for that); data shows peak RPM was about 29-30k. Taking into account the Voltage sag (from ~3.9 to 3.4 V/cell) in the data from the 160 kph run (see below), it means the motor was running at ~94% of it's no-load kv under ~150-160A load, I think that's pretty good performance, and I am a bit relieved I did not immediately run in to the 60k ERPM mother-of-all-efficiency-walls. I started with 4 degree timing (found some info from Neu on timing settings, although not sure whether 100% applicable to 56mm dia 6-poles: http://www.neumotors.com/NeuMotors/FAQs_%26_setup.html), but switched back to 0 after I experienced some glitches in throttle response, which I thought might caused by the timing; turned out it wasn't, but left the timing at 0 nevertheless, as it ran well and I already had all the speed I could handle.

@ Lizard: what timing did you run on your TP5840? Maybe wrong timing could be a factor in your resluts with it? Having seen how my TP5660 roars, I can't imagine yours not being able to top the TP4070, unless it's faulty.

What is a bit bothersome is the ripple Voltage under throttle ramp-up. I thought the amount of Current draw is what causes ripple, bit it seems it's more the combination of high draw, low duty-cycle... maybe I should practice ramping throttle up quicker, in order to get to higher duty-cycles faster, but then peak Currents will increase again, and there's only so much power the wheels can put on the road anyway before losing traction... so I'll also see if I can fit some more caps easily.

Gave the 2015 GRP tires another try (S3 compound this time), but they really suck balls and caused many near-misses and very high blood-pressure, in a bad way :), even hit my new camera tripod again, skating over the asphalt... luckily no major damage, but driving with those tires on a narrow street, is more scary than fun. Doing burnouts until they started to melt helped a bit to increase stickiness, but as soon as they got cold again it was back to drifting. Hope they go back to 2014 compound next year, or introduce some S0s.

Anyway, pics or it didn't happen:

New personal record:


Whe needs a body when you can have THIS instead:


160 kph run, with Torque Control on 5:


150 kph run, with TC on 4:


Loose gears:




Throttle glitches (sudden brake pulses out of nowhere...)


Settings:



Small video - not much but at least gives a bit of an impression of how it drives:
[YOUTUBE]-u0wiBMkxmE[/YOUTUBE]

Also realized I hadn't even bothered to check ESC temps:

[Lizard]

I had timing set to 'low' don't know how many degrees that is from tbe top of my head now, I think 5?

But are you sure it's looking good? I mean 48C at 17C ambient with a big fan on it and no hard driving is not that good or?

Almost 70C for the ESC also looks too high. Somebody on the offroad-cult forum said he had good results with removing the silicone stuff and the fabric and replacing it with this:
http://www.conrad.de/ce/de/product/1...-Kerafol-86600


Edit: Regarding the glitching: I have seen that on the XL2 too, but it was glitching in forward direction. Also something is weird with the ms values, it shows min 0.500 and max 2.500 sometimes, although your calibrated endpoints are around 1.000 and 2.000. Maybe an optical insulation thing like Vedder suggested for the VESC would help?

[ruudxd]

Regarding the screws, why not try red loctite?
Looks like everything is steel, so you can heat it up when you need to loosen the screws?
The previeus owner of my Muggy "glued" the pinion on the shaft with red loctite, and dunked the setscrew in it, but with a pencil flame it came lose pretty fast.

[Dr_T]

Originally Posted by Lizard

I had timing set to 'low' don't know how many degrees that is from tbe top of my head now, I think 5?

But are you sure it's looking good? I mean 48C at 17C ambient with a big fan on it and no hard driving is not that good or?

Almost 70C for the ESC also looks too high. Somebody on the offroad-cult forum said he had good results with removing the silicone stuff and the fabric and replacing it with this:
http://www.conrad.de/ce/de/product/1...-Kerafol-86600


Edit: Regarding the glitching: I have seen that on the XL2 too, but it was glitching in forward direction. Also something is weird with the ms values, it shows min 0.500 and max 2.500 sometimes, although your calibrated endpoints are around 1.000 and 2.000. Maybe an optical insulation thing like Vedder suggested for the VESC would help?

Might be a bit too early to draw conclusions, but I think the motor temps are what's to be expected and should give plenty head-room to push it more. The plot above with the ESC temps, shows 7 passes of 120-160 kph within 2 minutes, with the last two hitting 160 kph at 90k ERPM. My TP4070 runs at ~95% of the no-load RPM at 120A load, yielding 128 kmh in the truggy. The TP5660 runs at ~94% of the no-load RPM at ~150A load, yielding 160 kmh, so it can't be wasting a whole lot more power, if any.

XL2 temps are on the high side indeed, and I'm wondering whether that's mainly caused by the Current draw, or whether the high ripple Voltage is a factor in that too. Completely forgot to pay attention to temp of the caps.

Thanks for the info on the throttle glitching, didn't know it might be a common thing for the XL2. The glitches happen both forward and reverse, in the plot with the 150 kmh run there's three in forward direction, also exceeding what the Tx/Rx can put out as throttle range...

Originally Posted by ruudxd

Regarding the screws, why not try red loctite?
Looks like everything is steel, so you can heat it up when you need to loosen the screws?
The previeus owner of my Muggy "glued" the pinion on the shaft with red loctite, and dunked the setscrew in it, but with a pencil flame it came lose pretty fast.

No, the gears are plastic Ruud, the screws have very little plastic to hold on to and the plastic threads just stripped. We'll see, part of the problem is self-inflicted as I'm just too stubborn to get those expensive and heavy 8mm steel gears :).