Inside a 400amp 120v BLDC Kelly

Electric Motors and Controllers
User avatar
liveforphysics   100 GW

100 GW
Posts: 14049
Joined: Oct 29 2008 1:48am
Location: Santa Cruz, CA, USA

Inside a 400amp 120v BLDC Kelly

Post by liveforphysics » Jul 07 2011 9:40pm

It's a 42fet made from TO-220 package using this controller.

http://www.infineon.com/dgdl/IPP075N15N ... edead96fab


It's pretty much a guide on how to not to effectively parallel FETs to share current.

Image

Image

Image

Image
Each carcinogen vapor exposure includes a dice roll for cancer.

Each mutagen vapor exposure includes a dice roll for reproductive genetic defects in your children.

Each engine start sprays them into a shared atmosphere which includes beings not offered an opportunity to consent accepting these cancer experiences and defective genetics life experiences.

Every post is a free gift to the collective of minds composing the living bleeding edge of LEV development on our spaceship.

User avatar
bigmoose   100 MW

100 MW
Posts: 2664
Joined: Aug 06 2009 3:05pm
Location: North coast USA
Contact:

Re: Inside a 400amp 120v BLDC Kelly

Post by bigmoose » Jul 07 2011 9:49pm

A few small caps also, placed on the end of the rail. Not distributed across each bridge. What were they thinking?
Last edited by bigmoose on Aug 07 2011 7:22pm, edited 1 time in total.
"Live simply. Love generously. Care deeply. Speak kindly. Leave the rest to God..." all the best, Dave

User avatar
liveforphysics   100 GW

100 GW
Posts: 14049
Joined: Oct 29 2008 1:48am
Location: Santa Cruz, CA, USA

Re: Inside a 400amp 120v BLDC Kelly

Post by liveforphysics » Jul 07 2011 10:04pm

bigmoose wrote:A few small caps also, placed on the end of the rail. Not distributed across each bridge. What were they thinking?

They were thinking that they could call it a 400phase amp controller if they stuck 7 fets in parallel... Wrong! Lol
Each carcinogen vapor exposure includes a dice roll for cancer.

Each mutagen vapor exposure includes a dice roll for reproductive genetic defects in your children.

Each engine start sprays them into a shared atmosphere which includes beings not offered an opportunity to consent accepting these cancer experiences and defective genetics life experiences.

Every post is a free gift to the collective of minds composing the living bleeding edge of LEV development on our spaceship.

User avatar
Doctorbass   100 GW

100 GW
Posts: 7425
Joined: Apr 08 2007 2:24am
Location: Quebec, Canada East
Contact:

Re: Inside a 400amp 120v BLDC Kelly

Post by Doctorbass » Jul 07 2011 10:17pm

Hmmm A 400A kelly opened

Quite interesting, Thanks Luke

Now let's see inside a 250A to compare for fun :wink:

NOt that much difference in the fet number ratio.. the 250 look like 4 fets and the 400 look lilke 7 as you said...

Doc
Attachments
5614036613_a54a91e187_o_800x600.jpg
5614036613_a54a91e187_o_800x600.jpg (154.48 KiB) Viewed 5212 times
5614046899_90b19b7905_o_800x600.jpg
5614046899_90b19b7905_o_800x600.jpg (127.08 KiB) Viewed 5238 times
5614068159_f7b3588b26_o_800x600.jpg
5614068159_f7b3588b26_o_800x600.jpg (95.77 KiB) Viewed 5212 times
5614602740_57f0743e80_b_800x600.jpg
5614602740_57f0743e80_b_800x600.jpg (106.99 KiB) Viewed 5212 times
5614631382_2633a73c52_b_800x600.jpg
5614631382_2633a73c52_b_800x600.jpg (141.3 KiB) Viewed 5238 times
5614635710_25ff5930ca_o_800x600.jpg
5614635710_25ff5930ca_o_800x600.jpg (104.05 KiB) Viewed 5238 times
5614642874_f44acd1d81_o_800x600.jpg
5614642874_f44acd1d81_o_800x600.jpg (127.66 KiB) Viewed 5212 times
5614647558_6d2e54cd02_o_800x600.jpg
5614647558_6d2e54cd02_o_800x600.jpg (90.25 KiB) Viewed 5238 times
5614658306_ebdd681a54_o_800x600.jpg
5614658306_ebdd681a54_o_800x600.jpg (112.42 KiB) Viewed 5212 times
CURRENT PROJECT: 2WD duo MXUS/Max-E ebike 32kW
http://endless-sphere.com/forums/viewto ... =3&t=65764
-Fastest speed record 117 km/h on flat
-Fastest 1/4 mile@ 114km/h on flat and 16.316 sec
-Succeded to haul a 19200 pounds schoolbus!
113kmh Gianthttp://www.evalbum.com/3406
Mongoosehttp://www.evalbum.com/1947
YOUTUBE---https://m.youtube.com/user/Doctorbasss
http://twitter.com/DocbassMelancon
I speak FRENCH and english

User avatar
rhitee05   10 kW

10 kW
Posts: 841
Joined: Apr 07 2009 9:08pm
Location: Upstate SC

Re: Inside a 400amp 120v BLDC Kelly

Post by rhitee05 » Jul 07 2011 10:22pm

It looks like a tube of epoxy vomited all over the inside! Messy...

You might have a prayer of making that work if you carefully measured and matched each bank of FETs, and then threw a whole boxful of caps at it. Or you could just use the many hours that would take to create a better design... :-) You'd think the least they could do is use TO-247 FETs. I'm sure they could make up the cost difference if they added "Now with less explosions!" to their marketing materials.

Have you figured out exactly what the failure mode was?
Eric

Trek FX 7.3 hybrid - Cyclone 650W setup in process! Still...

User avatar
amberwolf   100 GW

100 GW
Posts: 25969
Joined: Aug 17 2009 6:43am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Inside a 400amp 120v BLDC Kelly

Post by amberwolf » Jul 07 2011 10:39pm

It looks like the way I tried "improving" my first brushed controller.... :lol:

flathill   100 kW

100 kW
Posts: 1273
Joined: Jul 06 2010 9:26pm

Re: Inside a 400amp 120v BLDC Kelly

Post by flathill » Jul 08 2011 1:36am

Ahh man you voided your warranty...Aha-ha-ha!

How did it die?

TO-220?? Really? They must be making a killing. No wonder they don't make them serviceable with the vibration goop and potted boards. I wonder if the newer black KHB series with beefier terminals is any better?

Still no luck with the Sevcon's ? :(

User avatar
neptronix   100 GW

100 GW
Posts: 13164
Joined: Jun 15 2010 5:56pm
Location: California refugee living in Utah, USA
Contact:

Re: Inside a 400amp 120v BLDC Kelly

Post by neptronix » Jul 08 2011 3:29am

MMMMMHM.....

You know i have read the complaint of these controllers getting too hot many times..
People wondering if they are defective etc..

I started thinking maybe they were overrated and started mentioning that.
Now i know for sure!

I think the amp numbers they're giving out are peak amp numbers and not nominal like the infineons are rated at.
IE it's a 200 amp controller, not a 400.

Nonetheless, that's BS.
My first major build: 8T MAC motor on a Trek 4500.
The new all-arounder: Leafmotor 1500w on a Turner O2 full suspension.

whatever   100 kW

100 kW
Posts: 1131
Joined: Jun 03 2010 2:16am

Re: Inside a 400amp 120v BLDC Kelly

Post by whatever » Jul 08 2011 4:25am

couple of questions:
1.the white goop is epoxy?? must be a good insulator too not short the fets
2. first link to infineon site, goes to to-220 mosfet data (assume they are the fets used in the kelly? not supposed to be a link to an infineon chip controller?)
3. is the board made by kelly or is it some chinese brand or similar?
thanks

User avatar
rhitee05   10 kW

10 kW
Posts: 841
Joined: Apr 07 2009 9:08pm
Location: Upstate SC

Re: Inside a 400amp 120v BLDC Kelly

Post by rhitee05 » Jul 08 2011 8:34am

neptronix wrote:I think the amp numbers they're giving out are peak amp numbers and not nominal like the infineons are rated at.
IE it's a 200 amp controller, not a 400.
That's really, really sad if they can only achieve 400/7 = 57A per FET as a PEAK rating. Design fail.
Eric

Trek FX 7.3 hybrid - Cyclone 650W setup in process! Still...

User avatar
liveforphysics   100 GW

100 GW
Posts: 14049
Joined: Oct 29 2008 1:48am
Location: Santa Cruz, CA, USA

Re: Inside a 400amp 120v BLDC Kelly

Post by liveforphysics » Jul 08 2011 12:27pm

rhitee05 wrote:
neptronix wrote:I think the amp numbers they're giving out are peak amp numbers and not nominal like the infineons are rated at.
IE it's a 200 amp controller, not a 400.
That's really, really sad if they can only achieve 400/7 = 57A per FET as a PEAK rating. Design fail.

Do you remember my thermal images of linear FET arrays being loaded? 8 monster TO-247 fets soldered directly by the drain tabs to a 1/4" x 1" solid copper bus on source and drain sides. Even those they were only like 8" away on a 1/4"x1" solid copper bar, the lead FET took the lions share of the current, as shown by seeing the leg temps in the IR just decreasing by like 10degC on each FET as it moved down the bus.

The way the bus is laid out, it wouldn't really matter if they had 42fets or 420fets, the FETs at the begining of the bus are going to get the lions share of the current density and begin the failing process.
Each carcinogen vapor exposure includes a dice roll for cancer.

Each mutagen vapor exposure includes a dice roll for reproductive genetic defects in your children.

Each engine start sprays them into a shared atmosphere which includes beings not offered an opportunity to consent accepting these cancer experiences and defective genetics life experiences.

Every post is a free gift to the collective of minds composing the living bleeding edge of LEV development on our spaceship.

User avatar
grindz145   100 MW

100 MW
Posts: 2928
Joined: Apr 07 2009 12:10pm
Location: Rochester NY
Contact:

Re: Inside a 400amp 120v BLDC Kelly

Post by grindz145 » Jul 08 2011 12:38pm

Cripes. I always had a good feeling about Kelly Controllers. Do all their controllers just use the same basic design and just scaled up by multiplying?

Anyway, Fail.

Thanks for sharing Luke! I love this.

User avatar
liveforphysics   100 GW

100 GW
Posts: 14049
Joined: Oct 29 2008 1:48am
Location: Santa Cruz, CA, USA

Re: Inside a 400amp 120v BLDC Kelly

Post by liveforphysics » Jul 08 2011 1:43pm

grindz145 wrote:Cripes. I always had a good feeling about Kelly Controllers. Do all their controllers just use the same basic design and just scaled up by multiplying?

Anyway, Fail.

Thanks for sharing Luke! I love this.

All that I've seen use this design.

We have one of this model comming:

http://kellycontroller.com/khb1480124-1 ... p-822.html

So, I will be able to tell you guys if they improved anything with the biggest BLDC they offer when it gets here.
Each carcinogen vapor exposure includes a dice roll for cancer.

Each mutagen vapor exposure includes a dice roll for reproductive genetic defects in your children.

Each engine start sprays them into a shared atmosphere which includes beings not offered an opportunity to consent accepting these cancer experiences and defective genetics life experiences.

Every post is a free gift to the collective of minds composing the living bleeding edge of LEV development on our spaceship.

User avatar
kentlim26   10 kW

10 kW
Posts: 653
Joined: Apr 04 2009 2:45pm
Location: Singapore
Contact:

Re: Inside a 400amp 120v BLDC Kelly

Post by kentlim26 » Jul 08 2011 3:31pm

thanks Luke,
all the while i had been thinking the kelly is the most branded controller. The buyers, we all don't know the actual true performace. Kelly controllers often mentioned themselves run by Top racing ebike, so i had nearly wanted to buy, i got to think twice before buying a controller from them now,. thank you very much. i had make a escape from buying now. their build of the controllers are often said to be very secrets also.

cheers
kentlim
Conhis hub motor run on April 2010, 48v 1000watts hub motor distance run 5800km, and Re~alive cyclone kit bought in year 2009, Bought a greentime controller 18fets- 4110 , 48v~ 90v , 3-speed mode. Latest top speed I got for my greentime controller 76.8km/h ! , yet to reach 80km/h, or I will never reached 80km/h ?

User avatar
rhitee05   10 kW

10 kW
Posts: 841
Joined: Apr 07 2009 9:08pm
Location: Upstate SC

Re: Inside a 400amp 120v BLDC Kelly

Post by rhitee05 » Jul 08 2011 3:39pm

liveforphysics wrote:Do you remember my thermal images of linear FET arrays being loaded? 8 monster TO-247 fets soldered directly by the drain tabs to a 1/4" x 1" solid copper bus on source and drain sides. Even those they were only like 8" away on a 1/4"x1" solid copper bar, the lead FET took the lions share of the current, as shown by seeing the leg temps in the IR just decreasing by like 10degC on each FET as it moved down the bus.

The way the bus is laid out, it wouldn't really matter if they had 42fets or 420fets, the FETs at the begining of the bus are going to get the lions share of the current density and begin the failing process.
You're right, to make that work they'd have to bring the battery buses in from the opposite side to equalize the path lengths. Seems like a little thing, but at those current levels it's enough. I wonder if it would be doable to attach the battery terminals to the opposite side of the busbar? Still wouldn't be a great design, but might help.
Eric

Trek FX 7.3 hybrid - Cyclone 650W setup in process! Still...

User avatar
neptronix   100 GW

100 GW
Posts: 13164
Joined: Jun 15 2010 5:56pm
Location: California refugee living in Utah, USA
Contact:

Re: Inside a 400amp 120v BLDC Kelly

Post by neptronix » Jul 08 2011 3:44pm

Luke, i did not see those thermal images. Looking at the controller again, i see exactly what you mean. I now fully understand how atrocious this design is.

Can't believe it. Am always shocked when such terrible engineering is well funded, marketed, and promoted so well. Seems to be the rule and not the exception.

I am now glad that i relented on buying one of these. They make a special controller for the MAC/BMC motors. They could not even begin to explain WHY it was special or different than the others. That was my first tip off that this may be a janky company.. oh and the negative reports that i mentioned.
My first major build: 8T MAC motor on a Trek 4500.
The new all-arounder: Leafmotor 1500w on a Turner O2 full suspension.

User avatar
liveforphysics   100 GW

100 GW
Posts: 14049
Joined: Oct 29 2008 1:48am
Location: Santa Cruz, CA, USA

Re: Inside a 400amp 120v BLDC Kelly

Post by liveforphysics » Jul 08 2011 4:11pm

neptronix wrote: Can't believe it. Am always shocked when such terrible engineering is well funded, marketed, and promoted so well. Seems to be the rule and not the exception.

Kelly isn't well funded bro, I don't think they even have their own engineers. It's a little shack in China assembling the through-hole parts by hand on boards that come in from a pick-n-place yard that handled the SMT assembly, then brushing over conformal coating and screwing them together in the extrusion. And a satellite office that answers phone calls from angry people who don't know how to get a working hall-combo, or want an RMA number to mail the controllers back that released the magic smoke. That is the whole company.
Each carcinogen vapor exposure includes a dice roll for cancer.

Each mutagen vapor exposure includes a dice roll for reproductive genetic defects in your children.

Each engine start sprays them into a shared atmosphere which includes beings not offered an opportunity to consent accepting these cancer experiences and defective genetics life experiences.

Every post is a free gift to the collective of minds composing the living bleeding edge of LEV development on our spaceship.

flathill   100 kW

100 kW
Posts: 1273
Joined: Jul 06 2010 9:26pm

Re: Inside a 400amp 120v BLDC Kelly

Post by flathill » Jul 08 2011 4:32pm

liveforphysics wrote: We have one of this model comming:

http://kellycontroller.com/khb1480124-1 ... p-822.html

So, I will be able to tell you guys if they improved anything with the biggest BLDC they offer when it gets here.
Awesome! The KHB has a beefier case (thicker endplates, 8 fan mounting holes on top, supposedly the new fet layout sinks heat to all sides of the case) in addition to the bigger terminals. Did you order the bolt on fans?

Also check out: Kelly KDH14651B post-mortem http://hr-ev.blogspot.com/2009/10/kelly ... ortem.html

He makes similar observations about the crappy layout with caps on one side and power terminals on the other. Hope the KHB is improved in this respect.

User avatar
neptronix   100 GW

100 GW
Posts: 13164
Joined: Jun 15 2010 5:56pm
Location: California refugee living in Utah, USA
Contact:

Re: Inside a 400amp 120v BLDC Kelly

Post by neptronix » Jul 08 2011 4:38pm

liveforphysics wrote:
neptronix wrote: Can't believe it. Am always shocked when such terrible engineering is well funded, marketed, and promoted so well. Seems to be the rule and not the exception.
Kelly isn't well funded bro, I don't think they even have their own engineers. It's a little shack in China assembling the through-hole parts by hand on boards that come in from a pick-n-place yard that handled the SMT assembly, then brushing over conformal coating and screwing them together in the extrusion. And a satellite office that answers phone calls from angry people who don't know how to get a working hall-combo, or want an RMA number to mail the controllers back that released the magic smoke. That is the whole company.
Wow.
I guess i was fooled by the shine, press releases, etc.
I need to get in business. If producing BS pays, i can only imagine what producing real products pays.

Or maybe it's the other way around. It probably is. :cry:
My first major build: 8T MAC motor on a Trek 4500.
The new all-arounder: Leafmotor 1500w on a Turner O2 full suspension.

User avatar
mauimart   100 W

100 W
Posts: 215
Joined: Nov 10 2009 2:08pm
Location: maui

Re: Inside a 400amp 120v BLDC Kelly

Post by mauimart » Jul 08 2011 6:36pm

Shit! I have one of these coming in the mail (KEB72800, 24V-72V, 300A, 8KW). Would not have ordered it if I knew that's what they looked like inside. :( The construction looks like crap. I am hoping it can do at least half the rated power on my Turnigy 80-100 (rewound 9 turns, 14AWG). For comparison I have attached some photos of the inside of a Curtis 1204 (brushed, 24-36V, 275A). Not that I'm a Curtis fan boy but It's a much cleaner design and layout, especially in terms of thermal consideration. Granted brushed controllers are easier to layout, not having all the isolation of the separate half-bridges, it's at least put together with some thought. Too bad they don't make a BLDC version...it would surely be better than the Kelly.
Martin
Attachments
P1050076.JPG
P1050076.JPG (98.04 KiB) Viewed 5537 times
P1050077.JPG
P1050077.JPG (77.67 KiB) Viewed 5541 times
P1050083.JPG
P1050083.JPG (81.58 KiB) Viewed 5541 times
P1050084.JPG
P1050084.JPG (72.9 KiB) Viewed 5541 times
P1050085.JPG
P1050085.JPG (94.78 KiB) Viewed 5541 times
P1050089.JPG
P1050089.JPG (71.3 KiB) Viewed 5537 times
P1050090.JPG
P1050090.JPG (92.11 KiB) Viewed 5537 times

User avatar
liveforphysics   100 GW

100 GW
Posts: 14049
Joined: Oct 29 2008 1:48am
Location: Santa Cruz, CA, USA

Re: Inside a 400amp 120v BLDC Kelly

Post by liveforphysics » Jul 08 2011 10:24pm

Yep, that curtis is way better than the Kelly layout. Caps down the middle is a huge difference in reliability alone, better thermal path, but same garbage linear fet array.

You know curtis does make big BLDC controllers. The bike my team won the last TTXGP formula 75 with used one.



Here is pics of the inside of a size4 sevcon.

Image

Image

Phase current sensors on the phase posts, which are located in the middle of the fet bank, which has caps on each side of it (almost like they were trying to make use of the FETs)

Image

Image

Image

Clips FTMFW! nuts and bolts are a super fail for holding FETs down (unless you have a bellville washer under each one
Image


Things look a bit different in a Sevcon. Same 42fet design, but now it's got decent current sharing, the caps are functional, and the FETs have an excellent thermal path to a solid easy to sink 1/2" chunk of aluminum.
Each carcinogen vapor exposure includes a dice roll for cancer.

Each mutagen vapor exposure includes a dice roll for reproductive genetic defects in your children.

Each engine start sprays them into a shared atmosphere which includes beings not offered an opportunity to consent accepting these cancer experiences and defective genetics life experiences.

Every post is a free gift to the collective of minds composing the living bleeding edge of LEV development on our spaceship.

User avatar
Arlo1   100 GW

100 GW
Posts: 8189
Joined: Apr 26 2009 10:36pm
Location: Nanaimo
Contact:

Re: Inside a 400amp 120v BLDC Kelly

Post by Arlo1 » Jul 09 2011 12:40am

Very nice. The screws with the plastic washers are gay I have melted lots of them lol! Sevcon is a nice piece of equipment.
My Leaf motor controller build. http://endless-sphere.com/forums/viewto ... 27#p963227
My YSR build http://www.youtube.com/watch?v=BRo8r5g4NBg
RC and most types of Lithium batteries you MUST know your individual cell voltages charging and discharging.
Don't keep them were you cant afford smoke or fire!
Never above 4.2v never below 2.7v EVER!!!
Support me on Patreon. https://www.patreon.com/user/posts?u=6842045
http://www.undergroundelectrics.ca/

User avatar
dave.com   10 mW

10 mW
Posts: 32
Joined: May 12 2009 8:33am
Location: SF Bay Area
Contact:

Re: Inside a 400amp 120v BLDC Kelly

Post by dave.com » Jul 10 2011 10:15am

Cool thread, and it strikes home, as I just plonked down some 600 bucks on a KEB72801. I just couldn't find anything else affordable that was buy-it-now-able with close to my target performance. I'm hoping it survives in _intermittent_ use at it's 150A continuous rating!

Interesting to see the Curtis cracked open, it reminds us that the limitations of Curtis's designs are sort of what spawned the Zilla so many years ago. I think Otmar originally started out modding 1231s or something similar. Though better than the Kelly, if you look at a Curtis apart on the bench and compare it to a Zilla, you begin to see how much work is really involved in building something capable of dealing with big amps/volts. Liquid cooling, massive billet copper cooling plate with machined in coolant channels, igbts, high grade bus caps, hairball interface, etc. Not to mention the thousands of hours spent training people to assemble the things properly. And of course testing. If you ever run into him, ask Otmar to tell a few of his plasma explosion stories. All this probably explains why there are no affordable, reliable, high power controllers out there for big brushless motors.

I do not know the man well, but I know enough not to ask him if he will help with a controller design (ok, I'm a liar, I did - but I won't again!). I think he is devoting his time to other loftier goals now ;+}

So who will the next Otmar be? While it's true he is an incredibly smart guy, he didn't start out as a degreed EE with 20 years experience designing high power stuff. He was a guy like one of us that was frustrated with the status quo. Through force of will and persistence, he taught himself the necessary skills to build, test, market and sell those magic green boxes.

User avatar
Arlo1   100 GW

100 GW
Posts: 8189
Joined: Apr 26 2009 10:36pm
Location: Nanaimo
Contact:

Re: Inside a 400amp 120v BLDC Kelly

Post by Arlo1 » Jul 10 2011 10:17am

Do you guys think its possible to mod a crystylite controller to put the chip from the Kelly in it? The reason I ask this is because my 24/36 fet bords have better cap locations!
My Leaf motor controller build. http://endless-sphere.com/forums/viewto ... 27#p963227
My YSR build http://www.youtube.com/watch?v=BRo8r5g4NBg
RC and most types of Lithium batteries you MUST know your individual cell voltages charging and discharging.
Don't keep them were you cant afford smoke or fire!
Never above 4.2v never below 2.7v EVER!!!
Support me on Patreon. https://www.patreon.com/user/posts?u=6842045
http://www.undergroundelectrics.ca/

salty9   1 kW

1 kW
Posts: 478
Joined: Aug 14 2009 10:14pm
Location: 42.6°N 122.71°W

Re: Inside a 400amp 120v BLDC Kelly

Post by salty9 » Jul 10 2011 10:32pm

Quote neptronix wrote:Can't believe it. Am always shocked when such terrible engineering is well funded, marketed, and promoted so well. Seems to be the rule and not the exception. Unquote

That business plan has worked pretty well for Microsoft so far. :oops:

Post Reply