GCinDC's Giant DH Comp

[gensem]

Dual RSP-1000 48v in series?
Not cheap but its reliable and will charge 100v around 25a

they are WAY MUCH HEAVIER compared to car1248 ones. And bigger in size. And much more expensive. So what is the point of using them?

2.5kw of charging power instead of 600w?
3,8kg only

edit: my mistake... it tough car1258 was the 600w BMS battery charger.

Im not sure about the diferences.

[andreym]

2.5kw of charging power instead of 600w?
3,8kg only

my car1248 weight only 1.3kg each and outputs 30a at 50volts = 1500w, so for 2 of them it will be 3kw in 2.6kg,
still 1.2kg less than overpriced RSP and 500w more powerful

[gensem]

2.5kw of charging power instead of 600w?
3,8kg only

my car1248 weight only 1.3kg each and outputs 30a at 50volts = 1500w, so for 2 of them it will be 3kw in 2.6kg,
still 1.2kg less than overpriced RSP and 500w more powerful

Do you know where to order the cart1248 with international shipping?

[andreym]

Do you know where to order the cart1248 with international shipping?

Unfortunately they are over on e-bay: http://www.ebay.com/itm/260847005833
and the seller told me he is not going to get more

[GCinDC]

[GCinDC]

aka how to sheer off the thread on disk...

[knoxie]

Oh No Greg you just aint having any luck? you need a bomber mate, that's it I'm ordering you one! on the house hmm suppose its a bit beyond liquid metals ability?

[GCinDC]

you need a bomber mate, that's it I'm ordering you one! on the house

address sent in PM

yeah, it's been a very destructive streak, that's for sure.

i've always intended to lace the M/C tire onto the hs3540, so let's just consider the old 9C just the test platform. it will go nicely on another bike, once i replace the side cover:

20120402_sheeredbrakeadapter.jpg (241.44 KiB) Viewed 276 times

the sheer failure is well known, right? but i'd expected the threads would get sheered, not that the entire thing would crack off.

20120402_evulsed9x7sidecover.jpg (230.41 KiB) Viewed 276 times

dbaker prolly had loctited the threaded adapter on. good stuff.

so moving along... i put in a good four hours last night rebuilding the DH Comp, starting with the upgraded hs3540.

the lyen ebike tester show that the blue hall wasn't working (solid red light)

but with a multimeter, i discovered a short between Blue and GND! Dang it! This then echoed in my mind...

Be sure to change out the hall wires too.

I could have ignored halls and installed sensorless controller, but would have preferred to use the sensored controller. so carefully removing the sidecover... can you see the frayed cloth? that was visible from outside and i suspected there was damage to the wires...

20120402_164732.jpg (245.21 KiB) Viewed 276 times

but with a magnifying glass i couldn't see any exposed wires:

20120402_inspectingmotorwires.jpg (228.62 KiB) Viewed 276 times

hmm. i checked the halls with the ebike tester again, and they all work - see the red lights going on and off when the magnet passes?

so the short must be in the wires, and if not now, maybe just when compressed in channel, so let's give the wires some more room:

i then taped up the wires where the cloth was torn, tossed the wheel back together, tested it on the bench and it sang beautifully! i ran it no load, WOT for several minutes, and watched as the motor temp went from 16C to 20C. woohoo!

i tossed it on the bike, and then lined the brake up perfectly with some washers - it rubbed with last motor and i couldn't stand any more NOISE. we learn. we learn...

i used a wire tie to clamp the wires away from the disk screws, just to make sure there wasn't any rub, and fired it up:

no longer singing, that's for sure. and when i put my ass on it and tested in the alley, you'll notice there's no mechanical rub under pedal power but with some throttle..:


yeah, we're used to this now...

no worries. just time to fire up the brand new sensorless controller. oops, gotta replace the connectors with bullets, and might as well watch some hulu:

20120402_changing_controller_connectors.jpg (232.46 KiB) Viewed 276 times

i tossed it on, and it tested fine, though a bit anemic... well, we'll see in the morning... i think it'll make it to work at least. i actually had trouble falling asleep cause i was excited to ride it.

and it did:

i had the old lipo in the pocket again, connected by alligator clips to power temp sensor, and damn, i never got above 64C! the controller was probably programmed with conservative settings so no idiots would burn up their motors:

eb312-settings.jpg (85.63 KiB) Viewed 276 times

Overall impressions of this build so far:
- brief jerky start from standstill
- surprising low torque during initial acceleration
- lackadaisical throttle response after coasting
- curious grumbing in middle range, wasn't sure if it was something in the motor, but it disappeared at higher speed
- eventually got up to max 38mph downhill, which was fresh after slowpoking around
- brakes feel good, and thankfully are now quiet due to proper alignment (lesson learned: tighten axle before aligning )
- need to reprogram to get regen?

so let's see if i can make it through the day without destroying anything - i might even like to get thru the week, but let's not temp fate.

ebikes are reliable, just as long as you're willing to work on them all through the night, if that's what it takes.

[grindz145]

OMG Greg, that looks so sweet!

It's all about sleepless nights when it comes to ebikes man.

My sensorless experience was MUCH worse, I'm glad to hear you're in much better shape!

[GCinDC]

it needs a cleanup. i didn't bother to wire-tie the wires, thinking i'd have to cut them off when something broke!

and i'm still trying to figure out how to bungee this bag. i forgot to mention that i thought the hall failure was actually due to a loose contact in the connector, that was yanked out because of the pressure of the bungee! turns out it didn't matter after i fixed it, but lesson learned... no more wrapping it around those wires!

and for my own reference, here's are the settings of my sensored controllers:

eb212sensored.jpg (92.28 KiB) Viewed 267 times

[grindz145]

That's some serious current for sensored operation. Be careful if you lose sync! Get off the throttle quick, or else bye bye mosfet

[GCinDC]

That's some serious current for sensored operation. Be careful if you lose sync! Get off the throttle quick, or else bye bye mosfet

i've had a pair of the eb212 controllers programmed with those settings since Nov 2010, and i can assure you, i've not been careful...

as far as the halls go, however, break out the marshmallows... (that said, the flat sided 9C in the JRH wheel that i ran all thru 2011 is still going..)

i might need to double-check the rshunt value, but on this morning's commute, the CA was reading close to 40A continuously up hills w/ sensorless that it used to read closer to 60A w/ sensored.

with the hs3540, you'd think i'd get a jump on a 9x7 @24s (especially if i sneak another 5s on board ) if ever sailorboy gets back...

ruff_vacation.jpg (46.45 KiB) Viewed 253 times

but now with this sensorless jobbie, i'm thinking i'll need to up the rated current to at least 60A...

[grindz145]

Greg, that current is fine with Halls, but once you "loose sync" on a sensorless controller, I would worry, very different scenario.



I missed it, who is sailorboy?

O wait is EB212 sensorless? I'm confused...

[GCinDC]

Greg, that current is fine with Halls, but once you "loose sync" on a sensorless controller, I would worry, very different scenario.

should i aim the gopro at the fet side of the controller to capture it?

O wait is EB212 sensorless? I'm confused...

eb212 sensored. eb312 sensorless. more at lyen.com

I missed it, who is sailorboy?

a local ebiker. i'd suggested we race before he went on vaca... to which he responded:

I'm charging my battery to 4.3V

[grindz145]

Haha, thanks for getting me up to speed

[GCinDC]

the love that keeps on coming...

preposterous.jpg (190.01 KiB) Viewed 504 times

[Skippic]

Yeah, and I accidentally charged them up to 4.46V. Lucky you it was raining ^^

[GCinDC]

5 miles thru the city at 7:30am w/o hardly stopping.

my_tracks_am_sensorless_commute.jpg (70.4 KiB) Viewed 474 times

the total time/moving time delta was just me dicking around in garage before heading out...

and to get up to max speed 42mph, i had to do pauld's death race tuck.
up that same hill last night, i pedalled hard to get to 30mph..

it's interesting riding w/o regen. i bet the motor (and controller) have a lot more time to cool down, cause when i'm running regen, the motor and controller are ALWAYS pushing amps, one way or the other...

no news on the fast chargers yet. i am somewhat concerned they won't shut off at voltage, and if the fan continues to run it may draw unwanted attention, but it does seem like an excellent value and form factor.

for now, i'm using the meanwell psu, and turnigy accuwhatever, which starts with an alarming fan whine:

should i hit it with a drop of oil?

oh yeah, i've not been balance charging, just lipo charging, but if i plug in the 5s balance connector, it seems to balance anyway... does that make sense? that just by virue of paralleling balance leads during charge that it would balance??? is so, what's the point of balance charging!?

[veloman]

I picked up a staple today too, went right through the Mr Tuffy tire liner.... It takes a bit of force to stab a really sharp knife through them. What are the odds that a staple can point straight up on the road? Then you hit it. It was a slow leak, didn't notice it till later while I was at home.

[GCinDC]

gosh this sensorless controller is anemic. it absolutely refuses to draw more than 3kw at any time. i can't even get the motor above 65C. how frustrating! so i just programmed it, upping the phase amps from 80 to 100 so we'll see about that...

meanwhile, the bike tire's got a bit more speed, but not much with this sensorless controller, and i do miss that sensation of the big M/C tire:


as you recall, i snapped off the threaded disk flange (by braking after jump), and while i'm doing repairs, i'd been wondering if i could fix the dish issue by...

...unscrewing each spoke (on the right side, that's laced on the rotor inside), bend the spoke enough to yank it out, and re-insert from outside of rotor, and tighten back up in nipple...?

well curiosity got the best of me, and unfortunately, in order to replace the spoke from the inside, it does need to be bent, just a bit...

20120410_212303.jpg (79.46 KiB) Viewed 360 times

otherwise it must go thru gap between the opposite side spokes and couldn't be bent back, certainly not w/o loosening all of them. the spoke can be bent back more or less (less) straight, but looks traumatized.

i'm not surprised, but i decided the main goal of this operation would be to gather some spoke data to rebuild the wheel from scratch.

step 1 was to let the air out of the tube, but i was amazed how stiff the tire was. i could easily believe it possible to ride on this tire completely flat!

but then i laid the wheel down and put a lot of pressure on the tire to 'break the seal'. then with clamps i was able to shift the tire/tube over enough to peek under at the tops of the nipples.

2012-04-10_20-45-20_HDR.jpg (215.69 KiB) Viewed 360 times

as i'd feared, my adjustments to undish it had overly loosened the one side (left one shown). notice how the spoke isn't threaded to the top/end of the nipple:

2012-04-10_20-44-22_HDR.jpg (119.97 KiB) Viewed 360 times

during an MC wheel spoke hunt, i'd found this instructive image:

Spokes7.jpg (69.9 KiB) Viewed 360 times

the obvious challenge with this wheel build is securing 12g (2.6mm) spokes with (5.5mm) butt ends in giant 8mm wheel holes:

20120410_205357.jpg (69.29 KiB) Viewed 360 times

dbaker accomplished it here with two washers, both of which bent to the shape of the hole:

20120410_205121.jpg (71.69 KiB) Viewed 360 times

i have no idea where he got the oval ones.

20120410_205212.jpg (83.87 KiB) Viewed 360 times

and now the data:

20120410_210416.jpg (57.68 KiB) Viewed 360 times

20120410_211055.jpg (98.73 KiB) Viewed 360 times

20120410_211159.jpg (74.7 KiB) Viewed 360 times

20120410_211233.jpg (79.85 KiB) Viewed 360 times

20120410_211003.jpg (84.96 KiB) Viewed 360 times

[binlagin]

What app are you running on your Andriod device to record those statistics?

[GCinDC]

My Tracks. Nothing fancy...

So I programmed the controller. (And this proves the chip is dead on my last one. )

Going to try it at 100A phase for a while and see how it goes. Some quick tests had it hitting 4kw regularly, and it felt better. Still weak off the line tho. And no regen yet. Looks like I need to set up an ebrake line.

[GCinDC]

Greg, that current is fine with Halls, but once you "loose sync" on a sensorless controller, I would worry, very different scenario.

Can you describe exactly what causes sync loss? You mentioned backing off of throttle quickly. Do you have personal experience with it, and what happens?

Considering how much stuff I've destroyed lately, I've wanted to go easy, so just upped the phase amps from 80 to 100 and the rated from 30 to 37.5 (do I recall correctly that these should stay in a 2.66:1 relationship?)

eb312_37.5A.jpg (81.78 KiB) Viewed 297 times

And I'm much happier with performance. As I said, it draws 4kw continuously which is a lot more useful in quickly getting around cars and pulling up to a max of 46mph.

It was cold this morning, and slightly downhill, but motor temp was only 65C! This really makes me wonder because my inbound commute w/ the sensored controller and monster wheel (on 9x7) that got up to 130C, no problem!

I may also be running cooler now because I don't have regen. I do want to set that up. Now that I've got hyrdaulic rear brakes, I can't use the electronic brake lever, so am thinking of dissecting one to get magnet and whatever to glue/screw/tape onto the hydraulic lever. Anyone got pics of such a setup? I'm still surprised how few people use regen...

[Skippic]

Forget regen, just get good brakes. For the brake switch you can use a hall or a reed switch. I'm interested if regen does heat up the motor as much as you say.

[sn0wchyld]

it does have a pretty notable warming effect... ive got my regen on a separate switch for that reason... so if i'm racing round i can just heat up the brakes, but if im on a long downhill or riding for range then the regens available when needed.

I use a micro switch, ziptied to the break leaver. that way, I can pull on the regen, and then pull a bit harder to add in the disk brake, or just pull the disk brake on its own. works like a charm.