Congenital Twin....Turnigy reconfigure test 1

*The quest for power*
Any one who followed my re-build of Aussie Jesters motor all the way through knows we had some ideas regarding power modifications & some of our current limitations regarding controlling these monster tourque machines.
After the 1st re-winding, AJ's motor and reading the #s it was aparent the the resistance was far to low to have any hope of keeping a controller alive at the ampdraw potentials we were creating......

this potential solution: re-wind from distributed LRK to a straight LRK....& then wind the remaining 6 teeth with another set of lrk windings....thus making this 2 motors on a single shaft.

Here are the specs stock # / rewind #s
Model 80/85-170kv
stock- 88 srtands 33gage 16turn DLRK terminated in delta/1 strand 14gage 10 turn terminated in wye x 2
copper cross secction: 17.9/20.7
resistance 23.14/16.5
res. wye 46.8/33.1
res. delta 15.6/11.0
copper weight 3-phases 177g/ 108.8 (times two sets=217 grams)
whats all this mean?
We have a 30% (I rounded the #s from drive calculator) increase in copper in this motor The reduction in turns has maintaind the original KV within a a few % point from stock I calculated 173 from the rewind.

I can now input power from 2-CC HV-160's I will let you guys do the math on peak potentials.
This is the 1st step of utilizing whats available to a mechanic like me to take it to the gassers on a sleeper build. Will be fiting this motor to a bike this evening & seeing how is all shakes out tomorrow afternoon. (getting dark now)
here is the photolog:
10 turns of 1.63 magnet wire

notice the epoxyed ends of the stator & wound with spectra fillimant for insulation

the madusa terminated in Wye. 1 set of black wires & a 2nd set of red


all assembled & wired up for some fun:

[youtube]_kU05jk1iis[/youtube]

whadya thinks????? can i get 60+ on the velodrome with this toy motor......this Hobo's going racing (wait till you see my hack on the ca120 :wink: )

Great work Thud, i'll keep my fingers crossed for you when you hook your 2xhv160s up, it will
either go like a bat out of hell or get very expensive very fast...lil like Lukes motor detornation
in his Civic recently LoL.. Only thing i am concerned about is a ESC losing sync, would be instant death for it
you are pioneer Thud, you push the envelope and blow stuff up so the rest us don't have to, well done mate
and thankyou.

Very much look forward to how it performs on the race bike! Best of luck

KiM

Why not put together 4 cheap Turnigy 100amp HV controllers. For $320 you could have a 400amp 12s controller. Compared to $280 for a CC Ice 160amp 12s controller.
Keep one brain board, with 4 power boards... Am I missing anything, that would make it not work...?

It should be possible... CC does it on their Phoenix line of controller...

HumboldtRc said:

Why not put together 4 cheap Turnigy 100amp HV controllers. For $320 you could have a 400amp 12s controller. Compared to $280 for a CC Ice 160amp 12s controller.
Keep one brain board, with 4 power boards... Am I missing anything, that would make it not work...?

It should be possible... CC does it on their Phoenix line of controller...

Click to expand...

I would guess keeping 4 controllers synced at all times is a good reason to go with the two...when/if
the controller looses sync it blows, Recumpence has done it in the past on his twin motored trike.


KiM

I have not seen the stacked CC controllers like that before now.......I assume they are running off 1 logic board?

hnnmmmm.That is a good question...the other is why would i stack 100amp controllers if i could stack 160's ......I'll be investigating this further. Thanks for the idea Humboldtrc.

the motor is wound to better the resistance levels of the stock wind by a tiny bit & aid in the thermal handling capacity of the wire.....its just an experiment to see if the senario is viable.

Thud said:

I have not seen the stacked CC controllers like that before now.......I assume they are running off 1 logic board?

hnnmmmm.That is a good question...the other is why would i stack 100amp controllers if i could stack 160's ......I'll be investigating this further. Thanks for the idea Humboldtrc.

the motor is wound to better the resistance levels of the stock wind by a tiny bit & aid in the thermal handling capacity of the wire.....its just an experiment to see if the senario is viable.

Click to expand...

Probably 50% of their big controllers are like that....






Any reason why we could't do the same with cheap esc's from HK. I want to put together 10 if possible, for a crazy 1000amp 50v esc. There is probably no way to cool the middle ones enough, when you start going over 3 or 4 boards stacked. So 4 is probably the limit of what you can stack and still have sufficient cooling to the middle boards, of the esc.

Anyone who knows about building esc's, your ideas would be a great help.

Is their a limit to what 1 logic board can control?

I would stack cheap $80 100amp controllers, not $280 160amp controllers. But thats my cheap ass. I love Hobbyking

there is a limit to how many FETs the drivers can handle. The cooling also becomes an issue when stacking too many controllers.


I like where this is going, certainly a good option for guys wanting to put more power down.

It apears to me that castle already does the stacking of controllers & just renames them for the higher amp potentals. I have never been inside a hv160 but i would guess there are 4 boards stacked to make it happen.... & the logic boards can only drive so amny fet's as JRH pointed out. Matt has expeirmented with using other logic boards from castle on his early builds & it didn't work so well back then....(boat contrller logic to run a pr hv110's fet setups).

Regarding the cheap-o controllers, I have a box full of them for experimentation.....they absolutly suck to program (get whopper hot durring programing) the programing cards look good but the one i bought for the hobby wing controllers works with my turnigy plush series controllers but not the mystery controllers.(wich i thought were made by hobbywing). & their sync abillity is embarasing compaired to a Castle creations. (you would have thought some one would have reverse engineered their start up sequence by now) I just have more confidence using the castles. Confidence is what I need blasting around the Dorais velodrome.

With the encryption that Castle uses, it isn't likely to have it hacked like the Jeti startup was years back. Reverse engineering, good luck on it. An engineer good enough to do that ought to be able to make his own design anyway :lol:

Anything that keeps current down while still delivering the power no brainer.
Many lower current controllers can function much better than a single giant one.
The only trick is, they really should be all sensored to ensure no sync issues, but hey, this is a bog step on the right path.

Outstanding work Thud!

Thud said:

Any one who followed my re-build of Assie Jesters motor

Click to expand...

Was that a typo or a slam? I can't tell with you guys anymore! :lol:

-JD

LOL Oatnet....I had to read that 19 times before i cought it.......totatal typo (this time ) & i did the math for multi-sensored controllers....we need to change the pole count to keep the timing correct on an ABC wind....with a single set of halls. After breaking the sound barrior in detriot this weekend I am all over this. Shure wish burtie would make a breakthrough on the timing advance programing. that would complete the package.

hi,

i found this site a month ago, and bookmarked it. this guy also wound an outrunner with dual windings. he said the motor has 4 times the power capabilities of an dLRK motor

is this true?

The consequences of this dual motor setup should be four times the maximum power of said standard dLRK motor. This is because each winding will be half the length, produce half the back-EMF, yet have half the phase resistance. Twice the current can flow, but the torque constant will be halved, and there are also twice as many windings now. However, because the back-EMF is also halved, the motor will tend to spin twice as fast.

Roughly speaking.

Click to expand...

nieles said:

hi,

i found this site a month ago, and bookmarked it. this guy also wound an outrunner with dual windings. he said the motor has 4 times the power capabilities of an dLRK motor

is this true?

The consequences of this dual motor setup should be four times the maximum power of said standard dLRK motor. This is because each winding will be half the length, produce half the back-EMF, yet have half the phase resistance. Twice the current can flow, but the torque constant will be halved, and there are also twice as many windings now. However, because the back-EMF is also halved, the motor will tend to spin twice as fast.

Roughly speaking.

Click to expand...

Click to expand...

Yes. If you don't re-wind to shift the KV back, that's exactly what happens.
However, that gives you a pair of coils to drive that are each twice as difficult for controllers to power.

Thud did this build with the purpose of making coils that are friendly to the ESC, so they share the burden of of the motor rather than increasing the burden on each controller.

Another example of world comunications.....there are lots of nerds fiddling with motors out there 8)
kool site nieles. interesting stuff.
I just hope someday to have an original idea!

johnrobholmes said:

... The cooling also becomes an issue when stacking too many controllers. .

Click to expand...

Hmm ? .. you could use the "marine" spec ones with water cooling jackets built in..??
..but then you have created a need for a heat exchanger and circulation system ..
and the marine ESC's are often higher amp rated to start with. ?

The "marine" controllers are better suited to low pole count motors (at least the castle creations units are) I have never played with one, But I know recumpence has with no Joy.

Just using the water jacketed fet coolers (?)...I would just submerge the whole controller in a coolent jacket for more optimal heat transfer. But I have been advised that this really wouldn't "save" fet's once they started cooking. Better i asume for these surface mount fets than the T0-220's I was going to run in a "wet" system.

LFP,(or Jeremy H if your looking) this is your area of expertise. Would a full imersion be of more bennefit to the SMT's over the T0-220's. Any one?

I had plans to flood a infineon controller, but put the idea aside on Jeremy's advice that the T0-220s wouldn't benifit enough from the flowing coolent. They roast internaly before the outside heat sinks come into play at the amp levels I am playing with.

Keep in mind I am pushing this stuff to the last 1% of the performance envelop. I will have some real data & video of this in action this saturday. Weather report looks phenominal & I have the "Post Apocylitic" racer ready for trimming. finishing the wireing tonight & let the good times roll. (hopefully its not rolling plumes of magic smoke) Thunderdrome here we come!
http://thunderdrome.com/
Back on topic:
Re-cap-I re-wound an 80/85 170 turnigy LRK (6 teeth)to match the performance of the stock DLRK. Then wound the remaining 6 teeth in the same LRK senario, to be independantly controlled. I have seen examples of this working from Axel on this board with his plywood axial motors, & now the with the link to the razor scooter modder....the limited bench testing has my confidence very High for this set up.
Things to come:
My next experiment will be getting a CA120 running from a pr. of infineon controllers off a single set of hall sensors. This will involve a modified poll count & possibly an independant timing ring for on the fly adjustment. (from POWER-to economy) Unless some one thinks its a total wast of time.....or untill Kingfish gets a controller built I cant break.(oh please have a break through, so I may just depleate batteries & not liberate my smoke in a plasma trail of broken dreams )

Submergion cooling would be a God send for the RC controllers. The little PCB's could get cooled this way, and they have a decent thermal connection to the FETs.

I wrote my concept for it in one of Matt's threads. Basically, put the controller in a little soda-can like heatsink, and throw in a mini brushless PC fan for circulation (they run fine in cooling fluids), and fill it with ATF or mobil-1 (auto tranny fluid, or the lightest mobil 1 oil you can find).

I agree, with the proviso that you may need to run a low viscosity cooling fluid to get good circulation between the stacked power boards, or better, increase the spacing between the boards with some stand-off's, just enough to get good flow. In general the SMT packaged FETs tend to be pretty reasonable at getting heat from the junction to the heatsink, by virtue of being soldered on to it. They still aren't brilliant in terms of junction to case thermal resistance (with the exception of some of the newer surface mount ones discussed elsewhere), but the higher thermal capacity and conductivity of pretty much any fluid other than air would still help a great deal.

Jeremy

Flooding a controller using the leaded parts would probably let you push the packages harder. If you can get heat out of the junction fast enough, the next bottleneck is that the package leads will want to melt. Flooding the whole controller would cool the leads and should let you do better than the usual ~75A TO-220 package limit.

Here is an update: (i wish i could call it a "progress" report) In a contest of 2 controllers the weaker one loses dramaticly


as Aussie Jest prophosised I experainaced a syinc issuse (after a 3 lap session around the lot with no issues). the ensuing fire was dramatic...

Not relly a test of power potentials..but the twin wind didn't blow my mind or anything. Didn't seem any differen't from the stock power out puts. finished the day witha stock 80/85-170 & the remaining castle controller held up fine. The bike road accetably & was very predictable in a slide. I didnt go too crazy as I was running no rear brakes & I wore out a new set of shimano V brake pads on the front 10 min into the rideing session. Discs will be required from here on out.

Consider this phase of experimentation finished.....1 test doesn't make much of a trend graph, but I am on to other things ellectricaly speaking. I have a set of 12fet infineons I am going to make work....stay tuned for more expirmentations soon.

I have to ask...are you going to bag that ESC up and send it back to Castle for warranty Thud?
Would really test their 'paid warranty' return policy wouldn't it ...

Positive side, the bike handles well, you used the 24in up front i assume as the 20in has the twin disks up front?
I am also guess it went lil quicker than the Thudster is letting on too :wink: seeing it takes a 500cc two stroke moto-X bike
to get the Thudster's excited hehe...

Hope the next phase of the experimentation goes well Thud, best of luck as always will be following with extreme
interest

KiM

Yeah, would def like to see the reaction when the Castle guys open your warranty return. :lol:

That seriously sucks Thud, I was hoping for a breakthrough. How did it go at the Thunderdome? I like the prizes for the female participants. LOL Maybe you should stick to trannys, and sell me one dammit. :wink:

Thud said:

I experainaced a syinc issuse (after a 3 lap session around the lot with no issues). the ensuing fire was dramatic...

Click to expand...

Proper job on that ESC Thud. -Much respect..... 8)
Burtie