[Billzilla]

First post, be kind.

I'm helping a friend of mine research building an electric hillclimb car. It would only have to run at full power for maybe five minutes per day in total so it could be charged overnight. Each run would be between about 30 seconds to about a minute, most of which would be at full power.

I have to put my hand up and say I don't know a lot (enough) about how to do this but one idea was to use something like 20 lithium batteries and have a controller as simple as a potentiometer connected to the throttle pedal, and as more throttle is applied it turns on solid-state relays in turn.
Something like one of these - http://www.aliexpress.com/item/LED-i...613675311.html
20 steps of throttle is plenty enough for such a car.
The problem with that is we'd need a DC brushed motor and from what little I know they aren't as light or efficient as a brushless motor.

I dabble a little with radio-controlled helicopters and they all use electronic speed controllers that run timing stuff to control the motors. I suspect such controllers for a much larger motor that we'd need for the racing car would cost a fair bit and the budget is tight on this one.
Would a compromise of a controller that only controls the timing and not the power output? That should be cheaper ....?

We'd be needing a motor somewhere around the 150 kW range ideally but maybe two smaller ones could be made to work.

Any suggestions?

[Duncan]

Hi Bill

Have a good read of the forum - everything you need is there
I am building a car I hope to use in some hillclimbs (and on a very small budget)

I am using an open source controller - look up OpenRevolt
a series of relays will not work at all well

http://www.diyelectriccar.com/forums...ghlight=duncan

[Qer]

Originally Posted by Billzilla

First post, be kind.

I'll try. At least I'll try to numb you down so you won't feel the killing blow when it comes.

Originally Posted by Billzilla

but one idea was to use something like 20 lithium batteries

Lithium has extremely low internal impedance. This matters in a very big way since a small mistake will make them turn on their master in an instant. Not like lead-acid that tends to be very forgiving, at least compared to Lithium.

Originally Posted by Billzilla

and have a controller as simple as a potentiometer connected to the throttle pedal

There's nothing simple over that, trust me. We spent quite some time just fine tuning the function of a potentiometer controlling a motor, it's probably a lot more tricky than you expect.

Originally Posted by Billzilla

and as more throttle is applied it turns on solid-state relays in turn.

First problem: The Solid State Relays you've picked are made for AC. They won't work with DC. See? Already not so simple.

Originally Posted by Billzilla

20 steps of throttle is plenty enough for such a car.

Errr. No.

Don't expect an EV to behave as you want it to. The motor current will depend on RPM, motor voltage, temperature, what granny had for breakfast and the chaos butterflies on the other side of the world. Trying to control motor current with a bunch of relays will best case lead to seriously crappy performance and worst case a burning wreck.

There's been a bunch people here in the forum that apparently has thought "How hard can it be?" and gone DIY-frenzy on the controller building. A few has succeeded but in most cases there's either reports of "setbacks" or "catastrophes" if they bother at all to report back when the controller went 4:th of July on them.

Even those that HAS succeeded has found out the hard way that it IS hard (especially if you want more than mediocre commuting performance) and I don't remember how many transistors Paul blew up before he got Revolt running. Come to think of it, I don't even remember how many transistors we have blown up in the process but it took a while before we worked out all the kinks of the S1.

Going for a relay layout might sound like it'd be simpler, but I doubt it. Long story short EVs didn't get useful performance until Lithium batteries and transistor based controllers. There has been people racing with relays but I doubt any of them used Lithium since you probably pretty much need the high input resistance of lead-acid to not blow the contactors up.

Originally Posted by Billzilla

The problem with that is we'd need a DC brushed motor and from what little I know they aren't as light or efficient as a brushless motor.

Um. I'd want a reference for that claim.

AFAIK all drag racing nuts still run brushed DC because it's more power per kilo than anything else. Efficiency tends to be shot when you start to go for racing performane and I seriously doubt it should even be a design goal. At least not if you want to have a sporting chance...

I know there were several EVs in Pikes Peak that ran AC-motors, but they had a budget for it too. Ie, HUGE! Lots of sponsor money.

Originally Posted by Billzilla

I suspect such controllers for a much larger motor that we'd need for the racing car would cost a fair bit and the budget is tight on this one.

"Racing" and "Tight budget" is a bit mutually exclusive. At least if you don't enjoy finishing last...

[major]

Originally Posted by Qer

I know there were several EVs in Pikes Peak that ran AC-motors, but they had a budget for it too. Ie, HUGE! Lots of sponsor money.

The Elias Anderson vehicle budget was on the same order or possibly less than your EVWest car. But yes, the Japanese entries appeared very high budget. I heard that Monster had a crew of 27 engineers with him

Pikes Peak is more like a road race uphill (upmountain). 30 second hill climbs are more like uphill drag races, therefore probably better suited for DC.

But Bill.....use a quality DC motor controller....for safety sake.

major

[Billzilla]

Originally Posted by Duncan

Hi Bill

Have a good read of the forum - everything you need is there
I am building a car I hope to use in some hillclimbs (and on a very small budget)

I am using an open source controller - look up OpenRevolt
a series of relays will not work at all well

http://www.diyelectriccar.com/forums...ghlight=duncan

Thanks.
I have a car similar to yours though it uses a two litre petrol engine of about 185 kW.



I assumed there's all the information I need here, but a few pointers in the right direction to reduce the confusion of what to look for would be handy.

Originally Posted by Qer

Lithium has extremely low internal impedance. This matters in a very big way since a small mistake will make them turn on their master in an instant. Not like lead-acid that tends to be very forgiving, at least compared to Lithium.

Has to be lithium for weight.

Originally Posted by Qer

There's nothing simple over that, trust me. We spent quite some time just fine tuning the function of a potentiometer controlling a motor, it's probably a lot more tricky than you expect.

Righto. It was just an idea I had that seemed to be the easiest way to do it, and something that we could build at home.

Originally Posted by Qer

First problem: The Solid State Relays you've picked are made for AC. They won't work with DC. See? Already not so simple.

I just picked that one as an example, I would have to work out the maximum current from each battery (or batteries) to see how big a DC relay I'd need.

Originally Posted by Qer

Don't expect an EV to behave as you want it to. The motor current will depend on RPM, motor voltage, temperature, what granny had for breakfast and the chaos butterflies on the other side of the world. Trying to control motor current with a bunch of relays will best case lead to seriously crappy performance and worst case a burning wreck.

Okay more research needed in this area then. I only picked '20' again as an example as 20 steps of throttle would be plenty for such a car. You typically drive one with either full throttle or nothing and rarely anything in between that. I figured ten steps might be a bit rough but 20 would be good enough.
Anyway, more research ....

Originally Posted by Qer

Even those that HAS succeeded has found out the hard way that it IS hard (especially if you want more than mediocre commuting performance)

Not interested in commuting performance, regen braking or anything like that. Just maximum current to the motor when you plant your right foot and a bit less with less throttle.

Originally Posted by Qer

Going for a relay layout might sound like it'd be simpler, but I doubt it. Long story short EVs didn't get useful performance until Lithium batteries and transistor based controllers. There has been people racing with relays but I doubt any of them used Lithium since you probably pretty much need the high input resistance of lead-acid to not blow the contactors up.

I picked solid-state relays as they have no moving parts. They are commonly used in performance cars to control electric water pumps and other reasonably high-current devices that need to be controlled for speed by duty cycle. Though the ones they normally use are only about 30 amps or so.

Originally Posted by Qer

Um. I'd want a reference for that claim.

Another example of stuff I am finding out about.

Originally Posted by Qer

AFAIK all drag racing nuts still run brushed DC because it's more power per kilo than anything else. Efficiency tends to be shot when you start to go for racing performane and I seriously doubt it should even be a design goal. At least not if you want to have a sporting chance...

That sounds reasonable then, a DC brushed motor would work for us okay.

Originally Posted by Qer

"Racing" and "Tight budget" is a bit mutually exclusive. At least if you don't enjoy finishing last...

Somewhat aware of that - I own four racing cars, two road cars, a rally car, etc.

Originally Posted by major

The Elias Anderson vehicle budget was on the same order or possibly less than your EVWest car. But yes, the Japanese entries appeared very high budget. I heard that Monster had a crew of 27 engineers with him

Pikes Peak is more like a road race uphill (upmountain). 30 second hill climbs are more like uphill drag races, therefore probably better suited for DC.

But Bill.....use a quality DC motor controller....for safety sake.

major

Had a quick look at a Zilla controller. (Hey, I like the name!!) and they aren't as expensive as I thought they'd be, so I'll have to do some sums and research to see how they would go for us. It's certainly seems to be the most effective and easiest way to go right now though.

[Billzilla]

FWIW some photos of the car.
There's one of it running on a track. The sidepods are likely to go as they aren't needed. I'll be fitting it with a big ground-effects floor that works far better and weighs only a fraction as much.

And a couple of photos of how of looks with all the bodywork off the chassis. It's far from a brilliant chassis and I'm making a lot of little changes to drastically improve the stiffness of it. But anyway keeping it on-topic there's a fair bit of room for batteries in the back.

[Tesseract]

Originally Posted by Billzilla

Had a quick look at a Zilla controller....

Qer - file under "No Good Deed Goes Unpunished"

[Qer]

Originally Posted by Tesseract

Qer - file under "No Good Deed Goes Unpunished"

I sense the irony is strong with this one...

[Billzilla]

Originally Posted by Tesseract

Qer - file under "No Good Deed Goes Unpunished"

They're NFG?

[gdirwin]

Zilla and Evnetics controllers are both top-notch and are both well-respected.

What is happening is that Qer and Tesseract are the brains behind the Evnetics controllers, whereas you mentioned the use of their competitor (Zilla). This is all fair game - they are both very active in this forum, and have been very helpful to all (even if you don't use their product).

I would recommend looking at both the Zilla and Evnetics products if you are looking at a quality dc motor controller.