Nevilleh's BMW

Post up a thread for your EV. Progress pics, description and assorted alliteration
User avatar
coulomb
Site Admin
Posts: 4323
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Nevilleh's BMW

Post by coulomb »

Sigh. Duplicate post. Edit, not quote.
Last edited by coulomb on Mon, 15 Feb 2010, 09:08, edited 1 time in total.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.
HeadsUp
Groupie
Posts: 265
Joined: Sat, 28 Feb 2009, 09:13
Real Name: Mark W
Location: Sydney

Nevilleh's BMW

Post by HeadsUp »

oh what fun

i lurve diagnostics


heat dissipation has been mentioned already ( dammit ) , but i would also be investigating at the fault being related to flash between components caused by insufficient electrical insulation and insufficient conductor thickness on the PC board , more so than a failure of a component , but certainly the heat transfer into the cooling element is a bit purplexing' too

wonder if they tried to copy somebody elses controller without having an ounce of comprehension of the design dynamics ?

NevilleH knows a fair bit about design , if they dont offer warranty and he can strip the controller it could have prospects for redesign and rebuild

just because components are black it doesnt mean they are all fried .

a new PC board and a few components , relocate heat sinks to allow a more direct path for removing heat from the active heat zones , as well as secondary cooling with airflow through the case itself might get you a workable unit
HeadsUp
Groupie
Posts: 265
Joined: Sat, 28 Feb 2009, 09:13
Real Name: Mark W
Location: Sydney

Nevilleh's BMW

Post by HeadsUp »


is there any components that have obvious damage , like the guts blown out of any mosfets ?

betting its on your list of things to check once you get through the grieving phase

Image
User avatar
Electrocycle
Senior Member
Posts: 985
Joined: Sun, 19 Oct 2008, 20:23
Real Name: Andrew
Location: Sydney
MSN: dumhed@dumhed.com
Contact:

Nevilleh's BMW

Post by Electrocycle »

coulomb wrote: Also, the heat spreader appears to be insanely thin. As well as that, the spreader seems to be 10 mm lower than the case. So maybe it doesn't even connect thermally to the case? Surely not. [Edit: maybe that's a trick of perspective; the PCB doesn't come to the edge of the controller.]

Maybe the busbars are supposed to connect thermally to the case somehow? Surely not.



the spreader looks to be thicker than the busbars, but is a separate part.
In the Curtis controllers it's a single U shaped extrusion.

The spreader sits against the case with an insulating sheet between them (electrical insulation - busbars / spreader are full voltage)

Fans blowing into heatsink fins are pretty normal - most of the airflow ends up travelling along the fins to a reasonable degree.
The Engine Whisperer - fixer of things
Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Nevilleh's BMW

Post by Nevilleh »

All valid comments. I won't pull it apart unless I have to ie no joy from the supplier, but it looks to me as if the copper cladding on the board gave way in a big way. Whether that was caused by a MOSFET melting I won't know until/if I get inside.
The heat sinking is pretty minimal.
Hopefully I might hear back from the supplier today.
Expensive business these electric cars!
User avatar
coulomb
Site Admin
Posts: 4323
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Nevilleh's BMW

Post by coulomb »

Electrocycle wrote: the spreader looks to be thicker than the busbars, but is a separate part.
Sigh. I really don't know DC contollers, it seems.

It looks like the thing I think is the bus bar is actually the heat spreader. How does it connect thermally to the case then?

The thing I thought was a heat spreader (over the capacitors) must just be some insulation or something, though it looks metallic in the photo.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.
Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Nevilleh's BMW

Post by Nevilleh »

Sorry my photos are not clearer. The black flappy thing is a piece of plastic. The MOSFETs seem to be bolted to Al pillars (the shiny bit where I rubbed the soot off with my finger) and these are bolted to the case itself.
I have received a reply to my plaintive email which says to send it back to the manufacturer and gives their address, so I am not going to delve into it any further. Hopefully, they will send a replacement that works!
Geez, a dc controller is not that hard, is it?
I wonder how one goes about suing a company in America?
User avatar
Johny
Senior Member
Posts: 3733
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Nevilleh's BMW

Post by Johny »

Nevilleh wrote:I wonder how one goes about suing a company in America?
Don't think that way yet. From my googling, LogiSystems appear to be replacing faulty controllers before they supply new orders and the post mid 2009 ones are going OK. I would include a letter explaining that this car is road-ready - not a long term project. They may move you up the queue if they like you.
HeadsUp
Groupie
Posts: 265
Joined: Sat, 28 Feb 2009, 09:13
Real Name: Mark W
Location: Sydney

Nevilleh's BMW

Post by HeadsUp »

Nevilleh wrote: ..... looks to me as if the copper cladding on the board gave way in a big way. Whether that was caused by a MOSFET melting I won't know until/if I get inside.
The heat sinking is pretty minimal.


one tiny component failure in a 4 cent part like a resistor can lead to a cascade failure in other bits.

the flash explosion of the PCB could have been just instantaneous release of the capacitor charge due to a failure of its normal path elsewhere.

tricky business doing diagnostics , you need to spend a couple of hours tracing back through components , sometimes there is too much destruction to get an accurate outcome , but if they have offered to take it back then the problem is all theirs , just a pity we couldnt learn from their stuff-up and incorporate it into our pool of knowledge.

Image
Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Nevilleh's BMW

Post by Nevilleh »

Well, we may never know as I have just packed it up and sent it off together with a very nice letter explaining the circumstances, etc.
$160 worth of postage. I do hope I get a quick response as my next move was going to be fitting the BMS to the car.
User avatar
acmotor
Senior Member
Posts: 3605
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Nevilleh's BMW

Post by acmotor »

Nevilleh, I think you need more than a 'repaired' controller. It needs to be a re-designed controller ! I'd be looking around for another supplier altogether.
Even if you could get that one replaced, you'd need to do a serious rebuild to have any chance.

Build your own if you like. You are quite up to it. But then it all takes time !!

BTW, don't lose heart. The 2km walk gives you plenty of thinking time !

The autopsy ?? forget it. Move on to a better controller.

Your power consumption at 100kmph 'uphill' seems quite normal. Where other EVs may slow down your emotor arrangement allows you to thunder on. But the controller must be up to it.

At 300A passing with maybe 1V drop across the controller that's 3kW of heat. That just defys physics in a little box. Yep, water cooled if you have to or just a serious heat sink/fins/airflow.

IMHO
iMiEV MY12     110,230km in pure Electric and loving it !
Tritium_James
Senior Member
Posts: 683
Joined: Wed, 04 Mar 2009, 17:15
Real Name: James Kennedy
Contact:

Nevilleh's BMW

Post by Tritium_James »

You won't get 3kW out of an air cooled box anywhere near that size and manage to keep the FET junction temp below acceptable limits.

At that sort of power density your options are pretty much water cooling, or a complex arrangement of heat pipes and large air cooling fin area.

What bugs me with these controllers (Curtis do the same thing) is that it doesn't even sense it's own temp and throttle itself back to protect itself. Seems a pretty obvious thing to do...
User avatar
Johny
Senior Member
Posts: 3733
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Nevilleh's BMW

Post by Johny »

acmotor wrote:At 300A passing with maybe 1V drop across the controller that's 3kW of heat.
Is it just me or did a decimal point slip here? More like 300W of heat (given the 1V assumption).
Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Nevilleh's BMW

Post by Nevilleh »

Yes, 300 watts is more like it! Ive since driven over the same piece of road in my "other" car and the fuel computer instantaneous reading drops from around 10 kms/litre to only 5 or so, so it must be a bit of an incline.
Thanks for the vote of confidence, acmotor, but not only does it take time, I have spent some $US1800 on this controller and I am loath to just write it off. But it is so frustrating to have the car all together and running on the road and be defeated by the incompetence of others. I don't mind my own incompetence holding me up, but that's why I pay others to do things!
Image
Perhaps I might have a wee look into the design of such things while I am waiting. That Zilla looks quite clever.
Tritium_James
Senior Member
Posts: 683
Joined: Wed, 04 Mar 2009, 17:15
Real Name: James Kennedy
Contact:

Nevilleh's BMW

Post by Tritium_James »

While acmotor has slipped a decimal point, it wouldn't surprise me in the slightest if the continuous power dissipation at 300A motor current was well over a kW, which is still going to be pushing things for a box like that.

As a comparison, in our 3-phase solarcar controller with 200V silicon, at 100Arms motor current we're losing 170W, and that controller efficiency is as good as it gets (synchronous rectification, fast switching, modern FETs, etc). Extrapolating our power loss curves to 300A motor at 150V bus would give 1.15kW dissipation.

Do they really rate it for 300A continuous?
User avatar
Johny
Senior Member
Posts: 3733
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Nevilleh's BMW

Post by Johny »

According to Chrisst907 on diyelectric car (I think he sells them on eBay).

Current/Time Ratings: All Models (LogiSystems)

550 amp - two minute rating 550 amp - 5 minute rating 400 amp -
1 hour rating 275 amps

750 amp - two minute rating 750 amp - 5 minute rating 600 amp -
1 hour rating 375 amp

1000 amp - two minute rating 1000 amp - 5 minute rating 800 amp -
1 hour rating 500 amp

I don't think this means all that much other than he was being pestered about real current ratings and finally responded with this.
This is the thread I found it on.
User avatar
acmotor
Senior Member
Posts: 3605
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Nevilleh's BMW

Post by acmotor »

Image
Yes, sort of typo as you note, but what I had included was the switching losses typical of such a controller without putting proper numbers on it.
Perhaps look at it another way..... when passing 43kW, if the controller is 94% efficient then heat loss is some 2.6kW.

I think the current quotes from Chrisst907 above still need to be shown on a SOA plot.
iMiEV MY12     110,230km in pure Electric and loving it !
User avatar
Johny
Senior Member
Posts: 3733
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Nevilleh's BMW

Post by Johny »

acmotor wrote:I think the current quotes from Chrisst907 above still need to be shown on a SOA plot.
Yes, I'm not disagreeing with you. When I first saw the fans blowing into an aluminium mini-wall I wondered.
Anyway, Neville, if you read that whole thread it's quite hopeful.
Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Nevilleh's BMW

Post by Nevilleh »

Thanks for all that valuable info, I will chase that thread down some more.
Just in passing, I looked up some Hexfets from IR and found the IRFP4768 which is a 250V device capable of carrying 66A at 100 deg C (93A at 25 deg) with an RDSS 0f 17.5 mR. I figure 16 of these in parallel could carry 1000A and the on resistance would be down to 1.09 mR, so if I stuck 300A through that combination the i-squared-r is only 98 watts. Not too hard to air cool that!
The no-man's land of 250V to 1000V seems to be littered with proponents of both Hexfets and IGBTs with no real clear winner as yet, although I couldn't find anything much in the Hexfet range above 300V.
Zilla use IGBTs and need water to cool them - not as low an on resistance as the Hexfet solution.

I just wanted to add that I haven't altered the controller settings, it is as supplied from the factory. Which is supposed to be "gentle".
The amount of time the car was running for was less than 5 minutes and I have the 1000A controller which has a 1 hour rating of 500A, if that thread is to be believed. No way it should have exceeded any of its "ratings".
Last edited by Nevilleh on Tue, 16 Feb 2010, 08:52, edited 1 time in total.
Tritium_James
Senior Member
Posts: 683
Joined: Wed, 04 Mar 2009, 17:15
Real Name: James Kennedy
Contact:

Nevilleh's BMW

Post by Tritium_James »

With 16 FETs in parallel your switching losses will be getting high, probably greater than the on-resistance losses. Then it's not synchronously rectified, so you've got the diode Vf * (1 - duty cycle) * Imotor in there too, if you're doing say 2/3 of top speed then you'd have around 1.5V * 33% * 300A = 150W. With that many FETs there's probably another 30W or so of FET capacitance * switching frequcency loss too.

So if you put those FETs in there (and quite nice parts they are, too!) then you'd be looking at around 380 - 400W loss. Older parts (even only a year or two older) could easily be double this loss. The latest generation of parts from IR are a big step up on what was available previously.
User avatar
acmotor
Senior Member
Posts: 3605
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Nevilleh's BMW

Post by acmotor »

Nice mosfets.
Keep in mind that even 1mR at 1000A is 1V drop (for the silicon junctions alone without busbars and pcb tracks). That's 1kW for on resistance and switching will likely be that much again.
The Zilla EHV IGBT is 1V at 1000A as well from memory and is not water cooled for any other reason than water is an effective (lazy?) way of extracting the considerable heat of FET or IGBT.

Beware extrapolating from data sheets for parallel devices. At 16x you'd best plan on only 50% continuous after load sharing and transients and remember the temperature gradient junction to ambient.

iMiEV MY12     110,230km in pure Electric and loving it !
Michael
Noobie
Posts: 4
Joined: Thu, 11 Mar 2010, 23:52
Real Name: Michael Gooi
Location: Malaysia
MSN: micchuan981@hotmail.com
Contact:

Nevilleh's BMW

Post by Michael »

Image Hi. Every one .
just like to get some help from all of you .
i have just convert my Alfa Romeo 146 to Electric.
the controller is only 48v 225A Curtis PMC
the car can only go as fast as 30 to 40 Km
Can any one tell me what type of power to go 110km
antiscab
Senior Member
Posts: 2700
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Nevilleh's BMW

Post by antiscab »

Hi Michael,

what year is your alfa 146?

in general i suggest 15kw/T cont for 100kmh, with 35kw/T availabe for hill climbing and overtaking (unless of course slowing down on 100kmh roads in your area doesn't result in traffic problems).

Matt
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells
User avatar
Speedily
Groupie
Posts: 123
Joined: Mon, 22 Dec 2008, 15:26
Real Name: Anthony
Location: Sydney
Contact:

Nevilleh's BMW

Post by Speedily »

Michael wrote: Image Hi. Every one .
just like to get some help from all of you .
i have just convert my Alfa Romeo 146 to Electric.
the controller is only 48v 225A Curtis PMC
the car can only go as fast as 30 to 40 Km
Can any one tell me what type of power to go 110km


What batteries have you got ?   
Michael
Noobie
Posts: 4
Joined: Thu, 11 Mar 2010, 23:52
Real Name: Michael Gooi
Location: Malaysia
MSN: micchuan981@hotmail.com
Contact:

Nevilleh's BMW

Post by Michael »

I m not sure of the cars year. but i think is 1998.
my web site alfaev.com
Post Reply