Help with drop in Torque

Open for any sort of non-technical discussion regarding EVs
Post Reply
lachlanmac
Groupie
Posts: 74
Joined: Sun, 16 Nov 2008, 05:56
Real Name: Lachlan Maclean
Location: Kallista Melb.

Help with drop in Torque

Post by lachlanmac »

Hi all,
I have a 91 Corolla conversion. 144v LiFePo, 8" Advanced DC motor and Curtiss 1231C controller. Just recently I have lost the ability to accelerate from low speed in 2nd gear. I have to now start in 1st up to about 35kph then change. Once the car is up to maybe 50kph in 2nd, the acceleration seems to be OK. I can still drive up our steep hill through the forest in 2nd at 60 - 70 kph as long as I get up to speed. On the flat, the slow speed acceleration seems less affected. It appears that the low rev peak in torque has been reduced.

Other forums suggested checking for poor connections which reduce the available voltage to the controller so it does not function properly. I cant find any heated joints. The voltage meter seems to still show similar results as in the past. On zero amps it might be 150V dropping to 120V when under the most load. The current will still go up to 180amps but doesn't seem to go higher now. I don't get that surge of acceleration and the current spike if I press the accelerator down harder. I've checked the accelerator pot. It changes from 0 - 5k ohms smoothly. The controller still starts in the lower frequency noisy stage then changes.

Is it possible for a controller to fail in a way to stop allowing the higher currents through but still function properly in other ways?

I'm wondering if I should clean the contactor terminals.

It seems to be the lack of higher current that is limiting the available torque.

Any ideas will be gratefully received. I can still drive to and from work but I need to be more strategic how I approach hills and starts etc.

Lachlan


User avatar
coulomb
Site Admin
Posts: 3833
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Help with drop in Torque

Post by coulomb »

Is the pack voltage sag worse when you start in second gear?

It could be something pack or BMS related. Most likely a battery strap gone higher resistance, I'd guess.

After a drive, you could run over the pack with a non contact thermometer, to see if there are any hot spots.
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.

lachlanmac
Groupie
Posts: 74
Joined: Sun, 16 Nov 2008, 05:56
Real Name: Lachlan Maclean
Location: Kallista Melb.

Help with drop in Torque

Post by lachlanmac »

Hi Mike,
Ive run over the battery terminals with my inbuilt contact thermometer (my finger). Nothing seemed hot, mostly slightly warm. Ill check the voltage drops in various gears.

I also have felt the various connectors to motor, controller etc. after a drive. Hoping to find a hot spot. No such luck. I have a big red button and cannot get to where the cables connect to it.

My thinking is that Torgue is related to how much current feeds the various magnetic fields. Something is limiting that current so its resistance - so I should see a voltage drop - or the controller.

The contactor has given some trouble (not disconnecting always) and is open in the engine bay so can get dirt in. Ill have a better look at that soon.

Thanks for the reply.

Lachlan

zeva
Senior Member
Posts: 422
Joined: Sat, 15 Dec 2007, 05:09
Real Name: Ian Hooper
Location: Australia
MSN: sigmunky@hotmail.com
Contact:

Help with drop in Torque

Post by zeva »

Normally, high resistance connections on the battery side of the controller have more of an effect at high speed - where power and hence battery current is higher - rather than low speed acceleration. At low speeds the battery-side currents are quite low compared with the motor currents, so in your case my guess is it's something inside the controller or "downstream" from it. (Or the potbox, but you have checked that.)

As you (and coulomb) correctly suggest, if the drop in performance is due to a loose terminal or poor contact somewhere, you'd likely see some serious signs of heating. Normally if you drive for more than a few minutes this way, something starts to smoke!

How old is the Curtis? The way current limiting works in those controllers (by measuring voltage drop across the MOSFETs), if a few MOSFETs have died the controller will continue to operate with a proportionately reduced motor current limit. If you can find someone who has a peak-hold 500A+ DC clamp meter, you could put it on one of the motor cables and go for a quick drive, to see if the controller is still able to deliver its rated 500A.
Ian Hooper
--
"Never doubt that the work of a small group of thoughtful, committed citizens can change the world." - Margaret Mead
http://www.zeva.com.au

User avatar
coulomb
Site Admin
Posts: 3833
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Help with drop in Torque

Post by coulomb »

lachlanmac wrote: Ive run over the battery terminals with my inbuilt contact thermometer (my finger).

Ah, the digital thermometer Image

Hopefully on the insulated parts of straps and so on, or with one hand behind your back and not touching the chassis.

An open contactor is also a good place to look for trouble.
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.

lachlanmac
Groupie
Posts: 74
Joined: Sun, 16 Nov 2008, 05:56
Real Name: Lachlan Maclean
Location: Kallista Melb.

Help with drop in Torque

Post by lachlanmac »

Hi Ian
The controller is about 5 years old. What you say makes sense wrt the performance Im getting. The other possibility is something wrong with the sensing of the pot.

I don't know anyone with a big clampometer but I don't think the controller will deliver the big amps as if I go flat to the boards at low speed there is now big jump in current and no acceleration to speak of.

Cheers

Lachlan

lachlanmac
Groupie
Posts: 74
Joined: Sun, 16 Nov 2008, 05:56
Real Name: Lachlan Maclean
Location: Kallista Melb.

Help with drop in Torque

Post by lachlanmac »

And better still, when one stops working, I have another ready to take its place.

lachlanmac
Groupie
Posts: 74
Joined: Sun, 16 Nov 2008, 05:56
Real Name: Lachlan Maclean
Location: Kallista Melb.

Help with drop in Torque

Post by lachlanmac »

Hi Ian again
I went out in the car and tried some stuff.
At standstill, if I floor the accelerator in 2nd, 3rd etc, the current doesn't get higher than 20 amps at first. If Im on the flat, the speed slowly builds and the current rises with the motor speed. There is no appreciable drop on the voltmeter. It reads what I would expect with a current drain of only 20amps. 1st gear allows me to get the motor speed up much fast so the amps go higher - maybe 100-140.

It really seems like something is severely limiting the current at low motor revs.

During summer, the controller has overheated a number of times and it has switched into its noisy mode with a loss of power. The Curtis site said that the controller is not harmed by this and didn't recommend stopping and waiting for it to cool down, so I haven't done that. I also have noticed on USA forums that they commonly have extra cooling (more fins or a fan) on the controller. I haven't seen Curtis recommending this.

In planning for the worst, what would you guess a mosfet replacement repair would cost?

You have made your own controller, haven't you? How has it been going?

Cheers

Lachlan

zeva
Senior Member
Posts: 422
Joined: Sat, 15 Dec 2007, 05:09
Real Name: Ian Hooper
Location: Australia
MSN: sigmunky@hotmail.com
Contact:

Help with drop in Torque

Post by zeva »

It does sound like it could be a loss of motor current capability in the controller. 5 years is a bit young for a Curtis to start packing it in (most seem to go for about a decade) but if it has been running hot, that may have aged it faster. A lot of electrical components have their lives halved for every 10 degree increase in operating temperature. Air-cooled controllers can all benefit from a little extra airflow and/or heatsinking.

I don't think we have anyone in Australia who's trained to repair them (could be mistaken?) so it'd probably have to go back to the US. Or if you're confident with repairing electronics, you could attempt a repair yourself. Either way, you'd want to rule out any other possibilities first! If you know someone else with a 1231 who wouldn't mind you trying theirs in your car briefly, that might be a definitive test.

The 1231 has about 20x IXTH50N20 MOSFETs at about $13ea, and replacing them would be quite labour intensive.. hard to know what Curtis would charge!

In case you're considering doing it yourself, here's a write-up I did a few years ago after disassembling one: http://zeva.com.au/Research/Curtis. If there are damaged MOSFETs, there's usually some physical evidence, e.g visibly damaged packages, burn marks in the area. If some have died, it's probably wise to replace the lot, as the others may be compromised too.

The other possibility is if a component on the logic board is compromised, it could be under-driving the power stage. Less likely IMHO because logic boards usually don't have things in parallel, so it's either "all or nothing". But perhaps the thermal limiting is stuck on for some reason - faulty temp sensor or something? Just speculating now..

Re my controller, the new design seems to be working well now after a very protracted development process! All that remains is waiting for the shipment of custom extruded housings. A couple of links below for your interest. I'm particularly excited about the 1000A model. I was doing some road testing with one in my RX7 today, it goes like a rocket! (Compared to the 600A controller I was testing previously..)

http://zeva.com.au/Products/index.php?product=106
http://zeva.com.au/Products/index.php?product=111
Ian Hooper
--
"Never doubt that the work of a small group of thoughtful, committed citizens can change the world." - Margaret Mead
http://www.zeva.com.au

lachlanmac
Groupie
Posts: 74
Joined: Sun, 16 Nov 2008, 05:56
Real Name: Lachlan Maclean
Location: Kallista Melb.

Help with drop in Torque

Post by lachlanmac »

Thanks again Ian for that very informative reply.
I read your article on the 1231. I think Im lucky to have mine last so long!
I read the earlier stuff again and will firstly look more closely downstream of the controller, at the connections, motor brushes, everything. Other electric car owners and drivers are few where I live and rarer still are ones with spare 1231s.
I can possibly repair myself if its easy to spot the damage. Im sure that Curtis would charge handsomely for a repair job as they charge for the product. I will also ask around if there are any motor controller repairers around here. Sometimes forklift people do that type of work.
The price of yours seems very fair.
Cheers

LAchlan

Peter C in Canberra
Senior Member
Posts: 496
Joined: Sun, 27 Jul 2008, 04:05
Real Name: Peter Campbell
Location: Canberra

Help with drop in Torque

Post by Peter C in Canberra »

I had exactly this issue with the same motor and controller. At speed the controller seemed to work fine but it took a bit less current and produced less torque when starting out. It seemed to be gradually getting worse. I used a clamp meter on the motor lead and saw that flat out I could not get more than 400A from the 500A rated controller. Now torque is proportional to the square of motor current so 400x400 is distinctly less than 500x500.
I replaced the Curtis with a Kelly rated to 144V/1000A but programable to the motor's rated limits of 120V/800A. The square of 800A is WAY more torque than the square of 500A and performance is WAY better than it ever was with the Curtis.
Daihatsu charade conversion 2009-18, Mitsubishi iMiEV 2013-2019, Holden Volt 2018-2019, Hyundai Kona 2019-present on the ACT's 100% renewable electricity.

lachlanmac
Groupie
Posts: 74
Joined: Sun, 16 Nov 2008, 05:56
Real Name: Lachlan Maclean
Location: Kallista Melb.

Help with drop in Torque

Post by lachlanmac »

I think you all maybe right.
The car is still driving. I can still go up our hill in 2nd at 60-70kph but its harder to get there and I have to start anywhere in first. The syncromesh is getting more of a workout and complaining a bit. Our other car just blew a headgasket / cracked head and needs about $2000. While the EV gets to and from even with a limp, its repair will wait. Im tempted by Ian's OZ built model. Nice price, local work. The current price of a Curtis from the Sydney supplier is $2500. Where is the value? They have certainly made their development costs by now, I suspect we need a bit of competition.
Thanks all
Lachlan

Peter C in Canberra
Senior Member
Posts: 496
Joined: Sun, 27 Jul 2008, 04:05
Real Name: Peter Campbell
Location: Canberra

Help with drop in Torque

Post by Peter C in Canberra »

My Kelly that was a huge step up in performance from the Curtis cost $1400 from EV Works. $1000 cheaper and way, way better. You won't believe the performance you have. I thought the battery or motor (same as yours) were the performance limit but it was the Curtis controller. The thing to be aware of is that you want a controller that can do a lot more current than the battery. You might think that the motor is only seeing 200A when the battery is producing that but it might actually be 600A at ⅓ the volts. The controller acts a bit like a transformer.
Daihatsu charade conversion 2009-18, Mitsubishi iMiEV 2013-2019, Holden Volt 2018-2019, Hyundai Kona 2019-present on the ACT's 100% renewable electricity.

lachlanmac
Groupie
Posts: 74
Joined: Sun, 16 Nov 2008, 05:56
Real Name: Lachlan Maclean
Location: Kallista Melb.

Help with drop in Torque

Post by lachlanmac »

Thanks Peter,
I slowly become more aware of what is going on electrically. From Ian's earlier posts about his Curtis experience too, it seems to be much older design technology. And their reputation for reliability must be questioned. I don't think its only Australia too, the USAians seem to be unhappy with the servicing costs of them.
I will have to look more at the Kelly as well.
Lachlan

Post Reply