[Richard Wood]

I did a roll down test the other day. Now I don't know if these are valid when done up a hill, so if anyone can advise. At 75km/h average I calculated 9kw (12hp) on the flat and 30-35kw (40-48hp) on the hill I'm targeting, which I would drive up for about 90 seconds. I had previously calculated the theoretical value of the hill alone at 25kw (34hp).
I'm wanting to work out whether an AC75 (or AC50) /curtis combination would be sufficient or whether I need to go with something water cooled that is more powerful but would add more weight (looks like both in cooling system and in motor) and take up more space under the bonnet.
75km/h would be just under 3000 rpm for my gearing with 1-1 through the transmission.
Richard

 

[Jesse67]

That makes sense, 25kw you calculated to take your car's weight up the hill plus 9kw measured just to overcome the aero drag and rolling resistance on the flat gives 34kw, close to what you measured with the coast down test on the hill.

The power required to go up the hill entirely depends on your vehicle weight so that value will change unless your converted car weighs the same as it does now. I believe the curtis controller is rated at 650A for 2 min which is 65kw electrical power at 100V, with probably about 55kw getting to the ground. That controller heats up before the AC50 motor does so if it can dish out enough power for the hill you'll be fine. Of course the lighter you can make your finished car the less power it'll take to get up the hill.

Jesse

 

[ruckus]

...75km/h would be just under 3000 rpm for my gearing with 1-1 through the transmission.
Richard

That is the beauty of a transmission. Just downshift to 3rd and go a little slower up the hill.

Put a monster heat sink on that controller or water-cool it if you can...

Cheers

 

[Richard Wood]

Thanks Jesse, under New Zealand regulations the car cannot weigh more than it does now - 1555kg GVMR so that is a limiting factor I'm working to.
Ruckus, going slower up the hill is not an option I am considering.
All things considered, I'm tending towards minimizing weight, using air cooling, heat sinks and just enough batteries to well cover range, ahead of adding a water cooling system, going for heavier controllers, motors and batteries that will make it more complex and harder to fit under the bonnet.
An exception may be to take the extra 30kg hit of the AC75 over the AC50.
That said, if a little weight investment in water cooling a controller will mean I make it up the hill at the speed I want, I need to look at it. The water cooled controllers I've seen are heavier and seem to be paired with water cooled motors that are both large and heavy (and will need to add water cooling system and water to the weight).
It's got me wondering about reasonably priced water cooled controllers paired with air cooled motors (at least as good as the AC50 though), any pointers in that direction would be great.
Richard

 

[tomofreno]

I did a roll down test the other day. Now I don't know if these are valid when done up a hill, so if anyone can advise. At 75km/h average I calculated 9kw (12hp) on the flat and 30-35kw (40-48hp) on the hill I'm targeting, which I would drive up for about 90 seconds. I had previously calculated the theoretical value of the hill alone at 25kw (34hp).
I'm wanting to work out whether an AC75 (or AC50) /curtis combination would be sufficient or whether I need to go with something water cooled that is more powerful but would add more weight (looks like both in cooling system and in motor) and take up more space under the bonnet.
75km/h would be just under 3000 rpm for my gearing with 1-1 through the transmission.
Richard

It is not clear to me how you performed and used the roll down test. Normally it is done on level ground (in both directions to take an average since most roads are not perfectly level) and then a curve fit to the data to estimate the rolling resistance and drag coefficient. Then to these forces you add the hill climbing force of mgsin(pi*incline angle/180), where m is vehicle mass, g is acceleration of gravity, pi is 3.1416, and the incline angle is given in degree. Is that what you did? What is the vehicle's actual weight? All you have given us is the GVW of 1555 kg, so we can't compare estimates using the actual weight to your test. You also haven't told us what ambient temperature you expect to operate in.

Imo the AC50 has too low of peak torque for a 1555 kg vehicle for acceptable acceleration, though it would do that little 2 km hill. If you go up it at 80 kph you are only on it for 1.5 minutes. The motor won't heat all that much in that time even at close to its peak output. I think I estimated around 5 C at 50kW output. Even triple that isn't a problem at even 45 C ambient. Controller temperature is more of an issue. Will you be driving for many miles before you hit this hill, so that motor/controller temperatures are already elevated? Might need water cooling for the controller in that case. The hill is fairly steep but too short to heat things up all that much. Curtis should be releasing their 144V controller sometime this summer, and HPEVS will be offering an AC76 motor to go with it, with higher peak torque than the AC50. I would wait to see what they offer rather than use the AC50.

 

[Jesse67]

Richard, hill climbing at speed is about power output, the larger AC 75 won't make any difference as the controller is still only able to produce a fixed amount of peak power. The AC 75 just produces more peak torque, but according to HPEVS it actually produces less peak hp, 67 vs 76 for the AC 50. It's the same controller for both motors and in both cases the controller will be what gets hot and starts limiting current, not the motor.

Now for the water cooling, you can add a water cooled chill plate to the Curtis controller, you don't need a controller that is initially water cooled to do that, this is what ruckus is talking about I believe. Lots of people have done this with the Curtis controller. A simpler option is a large aluminum heat sink attached to the bottom with a fan blowing across it, potentially less effective but a lot easier to do.

Is that 1555kg your curb weight or the gross vehicle weight rating?

 

[Duncan]

Now for the water cooling, you can add a water cooled chill plate to the Curtis controller, you don't need a controller that is initially water cooled to do that, this is what ruckus is talking about I believe. Lots of people have done this with the Curtis controller. A simpler option is a large aluminum heat sink attached to the bottom with a fan blowing across it, potentially less effective but a lot easier to do.

On my car the controller is mounted on the inner wing
So I cut a hole in the inner wing,
The controller mounts on the inside with the heat sink protruding through the hole into the area under the wing - there is a lot of air flow in that area!
Just make sure the tire can't hit it

The controller actually mounts onto an aluminium plate (6mm) that bolts over the hole, the heat sink is bigger than the hole (the bit that bolts on is smaller than the hole) and bolts to the aluminium plate

Heat sink
I had some copper sheet - cut it into rectangles, use some 6mm aluminium to sandwich it to the aluminium plate - bend the rectangles out into a flower and there you go!

 

[Jesse67]

That's certainly one way to go, the controller only really needs cooling while you're moving so if it gets enough airflow while you're moving you're good to go.

 

[Richard Wood]

Hi Tomofreno
I did the test on both level ground and up the hill. The level ground is a motorway running alongside the sea, which is the flattest ground I could find. Don't know about a curve fit, I was only really interested in the data between 70 and 80km/h but it seemed linear above and below that as well.
Yes, I added the theoretical calculation of the hill climb, as well as doing the test on the hill to corobroate. They came in 4-5kw apart.
The actual weight of my car is 1260 and I added 80kg for myself so on the day of the testing it was 1340. However, I've scaled all of these figures to 1555 to compensate as it seems the conservative thing to do, to work on the basis of the GVMR.
As for temperature. We don't have much extremes here - typically between 5 and 20 celcius. We do get very windy days though.
These figures are rough, they can't be anything else when you're measuring speed off an analogue speedo and time off a per second counter. However, the ball park result seems still useful.

Yes, I will have already driven 10km before hitting the hill and hadn't considered that the temperature might already be up a little. Thanks for that.
Richard

 

[Richard Wood]

Hi Jesse
Thanks for pointing out that power output depends on the controller. Have to decide whether to wait for the AC76 bigger controller.
Yep, I've see water cooling of Curtis controllers. It seems better to me to use a controller with it pre-designed in, but still, worth considering.
Yes the 1555 is the rating weight. Curb weight 1260kg
Richard

 

[tomofreno]

With that ambient temperature range I don't think you need concern yourself with overheating of the motor or controller if you use a finned heat sink and fan on the controller. The controller for the AC76 is not "bigger", if by that you mean more powerful. It has about the same peak power as the 1238R 650A controller for the AC75 due to its lower peak current of 500A, but the AC76 motor has higher peak torque which would give you more acceleration up to base speed. HPEVS says they are going to offer several motor windings, so you could choose the one that gives the best tradeoff in peak torque and base speed for your vehicle. You want the base speed high enough that you have peak torque at 80 kph for the hill. Above base speed torque falls off with increasing rpm.

 

[Jesse67]

From what I've heard the new controller will accept a higher voltage but will only output 500A so the peak power will be about the same, but I could be wrong on this. I would think at higher voltage, lower current the efficiency would be higher so there should be less heating of the controller though.

Have you found any AC water cooled controllers and motors that are available in the same cost range as the HPEVS offerings? I'd be interested as well, I've only ever seen the Scott drive on the greenstage website and they say sold out every time I look.

Depending on how big your battery pack is and how much weight you strip out you could end up lighter than your current curb weight. The AC 50 doesn't weigh that much, ~62kg, and 36 x 100Ah calb cells are about 125kg. A pack that size could provide the power you need but your range might only be 50 or 60km. Depends how much you need.

 

[Jesse67]

I think we wrote that at the same time Tom.... !