"peak power" - Trek 4300 20" Magic Pie build.

Well, zombiess has not blown his cromotor yet, so i think you'll be safe :lol:

As promised..

Mr. auraslip, here are the pics of the axle.. note.. the big bearing squeezes the wires pretty damn tight.. so what you're seeing is fairly maxed out..

pieaxle1.jpg


pieaxle2.jpg


And the Pirelli ml75 16x2.75" ..

pie_pirelli1.jpg


pie_pirelli2.jpg


This tire is boss as hell and fits the 20" pie perfectly!!!!
Now it is about 21" instead of 20"..
I have 4.25in of pedal clearance now :mrgreen: :mrgreen: :mrgreen:

It is so heavy and thick that i feel like i could ride over glass all day and never worry. I'm sure it will add some nasty rotational intertia... but it's helped the geometry problem out a hell of a lot.

I think i can go to a 24" wheel in front with a particularly fat tire on it.. like a cruiser 3.0" tire or something.

When my boss stops kicking my ass so hard, i will investigate.. put the motor back together.. do some test runs etc.
 
Mr. Neptronix - THANK you. I assumed you grinded that axle down to help fit more wires?

In other news - I'm building my H series into a 24" rim.


$20 for the rim. http://www.chainreactioncycles.com/Models.aspx?ModelID=60742

$35 for custom cut spokes from ebike.ca (this is an incredible deal!) Just tell them the ERD of rim and the appropriate measurement of the hub.

Get a new disc hub for $50, and slap on a 2.5" cyclops for $20 and you should be pretty good. Handling will probably still be a lil' wonky since IIRC that setup is 25" and only lowers the front by an 1". Although it probably helps that you aren't running a suspension fork. If the frame is setup for a 100mm travel fork and you're using a solid fork made for 80mm bikes it'll help with handling issues. Hrm. I wonder if you could use an even shorter fork to get the handling right. Maybe a fork made for a 24" bike frame?

Your bike is starting to look pretty solid! Pretty much all it needs is a wooden battery box and then it'd be something I'd build ;)
 
I ground it down indeed. About half a millimeter to just barely squeeze the extra temp sensor and fan power wires through. May be able to fit 13g or 14g phase through it, if i take the temp sensor and fan wires out. Dunno.
It's too thin to grind on much more. I've seen the broken axles on here... :p

H series in a 24" rim should be pretty good. I've heard of 22" rims though. Finding tires is probably a pain though. Nice middle ground however that could be doable with a solid fork or very short sus. fork on a bike designed for 26".

That's a damn good deal on a 24" rim.. hmm :)

The local used bike store should have something i can line up for a test fit. They've wanted to see this bike for a while anyway ( got a few of them hooked on the idea of an eBike by letting them ride the MAC build on ~2500w.. i love spreading this ebike disease ;) )
 
Now i need high temperature capable zip ties to put this motor back together.

All local vendors have zip ties that melt around 85C.
Nobody knows any vendors other than a company a state over who can custom make me a minimum quantity of 1000 for around $500.

...and this is what sucks about handing all of our manufacturing over to China. A Chinese person could probably buy these at the supermarket. We have to order them over the internet.

Delayed again i guess.
 
zombiess said:
Could you use stainless steel zip ties... if you melt them chances are you are on fire and they aren't much of a concern.

true, but they are in a motor environment. I'm thinking of what could go wrong here if metal pieces got loose inside. You know how metal heats up, expands etc... can't say i trust it.

heatsrhink is a better idea..

Some high temp zip ties would be best though.. crazy expensive.. but no worries.
 
I use these inside my hub motor. I don't think they'll fail unless they completely melt. There isn't exactly much reason for them to fail considering they aren't load bearing.

http://www.amazon.com/Thomas-Betts-TY525MX-Cable-Tie/dp/tech-data/B001DEHZ14

They were recommend in the zip tie thread awhile back.
 
Melting is what worries me. 11 miles of 7% grade is going to make heat. I have no idea how much. I can only learn from a 1/3rd scale model of the hill that i can legally ride up.

Most zip ties are rated to 185f. That's only 85C.
The expensive ones can deal with much more heat.

So the challenge is finding some that can handle at least 100c/220f or so. I believe the wiring insulation fails at that point.

Those ties look good but i have something similar already.
 
Well, i spent >2 hours researching and trying to find high temp zip ties to replace the ones i took out of the motor. Lol.

Consensus is that they're really ****ing expensive.

But i found some old stock thomas-betts on eBay.. $30 for 500 high-temp ( 105C ) ty-rap. 100C is good enough. That's the point where the motor may look a little melty. I'm a happy camper now.
 
Action news update:

The bike is finally back together. The temp sensor works ( woot ) and the rear pirelli is a dream to ride on! totally takes the hit out of the bumps - i don't feel any vibration in the road. It's wonderful, makes the hardtail bearable for sure :)

The geometry is now really good for riding, but a 24" front wheel will really make it perfect. I already test fitted a 24" and am looking to get a 24" disc brake wheel.

wheeldifference.jpg


Top speed of the motor has gone up a mph or two :shock: Now fully charged at 20S, i am running 32-33mph.
This is too high of a speed for the hill climb, so i may drop down to about 16-18S.

The temp sensor is working, and the 60T -> 11t gearing is perfect for pedaling from 25-35mph.
The 145mm cranks aren't as bad as i thought they'd be for pedaling. ( normally i use 175mm cranks!! )

So, the hill tests will begin soon.
I will climb the 1/3rd scale version of pike's peak and see how it goes.
If i hit over 110c, i will drill cooling holes and try again.
If i hit over 110c still, i will add the fans.

If that doesn't work, i am just strapping an eTek to it and calling it a day :lol:
 
Looking good Dave :D

I really like that Pirelli !

Did you use the LM35 temp sensor ? I have on in the mac now and it's will really help protect it.

I think you are almost ready to win the Pike's Peak race! :mrgreen:

Some test rides now ?
 
auraslip, i had a mental slip. i meant 110c. fixed post.
 
BTW, thought i'd do a little document on how i did what i did..

tempsensor.jpg


temp sensor mod

precharge.gif


precharge resistor
 
OK, so the temp sensor is freaking great. The hand test is DEFINITELY useless. I cannot tell the difference between 50C and 95C with my hand :shock: . I didn't do this sooner... why? :D

I took the motor up my local 0.5 mile version of pike's peak, and another decent hill climb that's a bit less intense, but not the official test hill.

Got it up to almost 95c. The hill ended, so i couldn't test further :|.

Regen started kicking in at 76v. I guess the high end cutoff was set to that voltage. The regen was freaking strong, and sucked up to 1200w. Guess who's axle nuts got loose? yeah.... :evil:

I guess i will have to get to programming this controller & turn the regen down if possible.
 
OK, i finally got to programming the controller :D

Tuned the phase amps down some.. now the wheelie problem has improved significantly, tho it is less of a torque monster. I guess i will have to find the perfect balance ratio phase amps and battery.. IMHO, each motor seems to have it's own ideal ratios, and it seems like at high power levels, tuning it down from 3.0x the battery amp is always needed.

Slip charge got turned off ( bit too strong, unexpected, also prevents freewheeling and that sucks ).
EBS got tuned down to the '0' setting.

Now she takes in 10 amps instead of 18 amps, and the axle nuts on a test drive finally stayed seated.

DAMN.. 700-800w of regen is freaking strong. I think the low speed wind motor makes that 10 amps so much more powerful.. but this is the first time i have experienced regen on anything.. now i need to do the R12 mod.
 
Looking like a complete bike ready to ride now, eh?

I hope the testing goes well.

Do any of you guys know what is a safe temp limit for the Mac? The highest I've seen after pulling my bike indoors and checking the temp sensor wires with my voltmeter is 76-80 C. I heard something about magnets losing their power above that. I am going to experiment with putting some ice wraps around the hub, during the 2nd half of my trip during the hot lunch hour I take, where it's uphill coming back. We've still not gotten near the 105 degree days we will be seeing all summer, so I want to be on the careful side, especially once I go to 1500w range. (300w more than now).

Nep - if I was you I'd look into thermal-gluing tiny heat sinks all over your Pie. You could probably double the surface area.
 
Yeah it's pretty complete for the most part. All the big kinks have been worked out now :)

I dunno about your MAC question. 100C, maybe a bit more is usually a nice safe zone. The magnet enamel on most motors seems to be able to tolerate 150c for some time, higher for a very short period of time.

I think the limit of the magnets is 125-150c.

I think your ICE idea is going to fling things around. The MAC seems to saturate pretty bad around 3.5-4kW, mine heated up very rapidly on 4kW and the gears died quick, remember..

It's not a good cooling design at all, so it has it's limitations.

As for the pie, the cover sheds heat really well, and the aluminum rim acts as a heatsink as well. You could not ask for a better arrangement :). I am amazed at how much heat it can shed continuously, and how fast that heat is shed.

Soon i will run the big 2kW lawnmower version of the MAC motor, chain driven with an approx. 5:1 gear reduction, and no planetary gear reduction, so it can be vented ;].. and only 11lb. If it can put out more power than the magic pie, i will use it instead.

But i doubt it. The pie rocks and i can't wait to throw it up the 1/3rd scale pike's peak hill. I just need to get regen working at high volts so i don't die on the way down :lol:
 
Hi Dave,

Sounds like you are almost there, I would watch the regen if the pie is already pretty warm because I toasted the halls using regen by not giving the motor a chance to cool down. It was roasting hot and the smell of it in the car the whole way home from the mountain climb was bad. But it survived and I didn't bother to fix the halls. The magnets didn't loose any strength. And the windings were perfect looking.

I think most of my problem was the pie bogging down, I was dumping 3kw into it and it was bogging down, always a bad sign. But sometimes I wonder if I was running higher voltage would it have allowed the pie to climb faster spinning more efficient or because of more voltage means more watts would I have just heated it up faster ? I would like to test it some day. But in your 20" wheel it should have a much easier time!

VELOMAN

Paul said you should back off the power before 150 Celsius, I have the temp sensor now and I got back home from a 7 mile spin and it said 1.6 volts on the multimeter, 160 celsius, but the case was just warm, so it just goes to prove you can never tell with your hand especially with geared hubs!

Now it was a very windy day and the max I could do was 30 mph without getting blown into the ditch and it was pulling 3kw for longer than I would have liked, I don't know if the lm35 can read above that ? But the wind was gusting about 50 mph so that is a hell of a lot.

Paul also said that I shouldn't worry because I never had issues with heat before and there is no reason I should this time either and I think he is right. One time I did get the cover very hot so god only knows what the temp was inside then!
 
Scorpion: i can't seem to find any correlation between regen and motor heating. But high temps do take a while to cool down. A sustained high temp could very slowly kill the halls.

I saw the temp going down as i held the regen on.. but this was not on a particularly steep downhill.
BUT i am using a weak regen setting, setting '0', which takes in only 10 amps. On the 20" wheel, it is mega overkill.



So i hooked 25S / 60A power up to the motor and took it up my nearby test hill, which is not too steep, but very sustained.

[youtube]dWJSYGAqm0A[/youtube]

I got the motor up to 120c. Finally, this got the case hot, at 3000-3500w constant! NOW i can start testing the cooling mods :)

p.s. the cycle analyst speed is not correct - it is off around 10%. I was doing up this hill at 35-38mph. Early in the video, i am being nice to the throttle because the road is bumpy as well.
Will fix up the camera angles etc later on.


Tomorrow, the motor gets drilled and i will see how big of an improvement that provides.
 
Back
Top