Mazda MX-6 build thread (EV conversion)

[thingstodo]

Originally Posted by Astro

Somebody mentioned the pre-charge contactor circuit. I think i will put the emergency cut outs in the earth return that will be used by both the pre-charge and main contactor coils.
I don't think the pre-charge resistor would last long at full current and so wouldn't be a huge issue in a contactor fail on scenario. The pre-charge resistor would act like a fuse at full amps.

The pre-charge resistor limits current into the controller, so it won't be a fuse. The controller gets just enough power to run the electronics, but not to turn the motor.

My precharge gets quite warm if I forget to pull in the main contactor while I'm testing on the bench. I haven't cooked one yet!

[thingstodo]

Originally Posted by Astro

Vacuum pump if needed for brake assist.
Electric power steering pump unless you use the original pump driven off the rear shaft of the drive motor or have a manual steering rack.

Yes - I did miss those. I have the power steering pump and was thinking about using a separate 12V motor to run it. But it ends up being QUITE heavy. I'll find out how badly I need the vacuum pump when I get it rolling in my driveway. If my wife cannot get the brakes to function, I will add the vacuum pump.

You listed 50lbs for liquid cooling.
Is this for cooling the controller?
This seems like a lot to me. Of course i have no idea what would be involved in cooling a controller but i would have thought a few of litres of coolant at the most with a nice fan assisted aluminium radiator to cool it down. Like the after market ones they sell to cool transmission fluid.

Yes, the controller is liquid cooled. I kept the original rad with the original fan. There's a few liters of fluid, the pump, and some piping. The controller is happy up to about 55C so I was considering using the coolant to feed the original heater core. I'd still need a ceramic element to keep the windows clear until the controller warmed up ... and I'm not sure how the plumbing would work. I may need to add a shutter on the rad for cooler weather.

I should be high on the 50 lbs. I forgot, until I started typing this message, that the rad and fan and fill bottle were not weighed coming out of the car.

IGBT controllers, which is most controllers that allow voltages above 200V, have a voltage drop across the transistor. I've seen specs as low as 0.6V but they seem to be around 0.8 - 1.0 normally. When you get above 1/3 to 1/2 of the rated amps (which should be into your design factor or just for acceleration) the voltage drop increases - 1.0V - 1.8V. Since Power = Volts * Amps, a 1000 amp start at 1.8V drop is 1800W. Getting 1800W away from your controller is difficult. So the heat sink is designed to absorb much of that for a few seconds of acceleration and then the liquid cooling takes it away while you are coasting/cruising to the next light.

MOSFETS have a much lower voltage drop, but they can't pass as much current. So the MOSFET controllers have a bunch of MOSFETs in parallel to allow 1000 amps or 500 amps. But if you have 1000A at 0.2V, that's 200W and an air-cooled heat sink can do it (you still need to fan cool, and the heat sink has to be big, but you can do it)

I'm planning to use a mineral-oil coolant, like they use in large industrial transformers. It is not conductive, so if it leaks my controller won't be damaged. The coolant WILL BURN, but if the temperature is that high, my cables and all of the plastic under the hood is ALREADY on fire.

[thingstodo]

Originally Posted by Astro

But then i GPS logged my commute and found that i would need much less battery capacity than i originally guesstimated. 20 kilos less, so i am back to factory weight again.
My commute is not a long distance but lots of waiting at traffic lights.

Wow - you seem to have a strong handle on what everything weighs and what battery capacity you need.

What does you car weigh and what are you projecting for w-h/km or w-h/mile? Does that include some 'fun' driving?

I'm expecting the Mazda will be around 350 w-h/mile at first. When the tires need to be replaced, low rolling resistance tires should drop that a bit. I think one rear wheel bearing need to be replaced, which should help as well. The emergency brakes are dragging a bit, so fixing that will help as well. After that is closing off the radiator space up front (if the cooling system can handle it), adding to the hood to streamline the wipers, perhaps eventually a belly pan.

At 350 w-h/mile, I just make my commute with no safety factor. I will charge a bit at work, just 110V, and keep the batteries warm so that they perform well. But it's not much of a safety factor.

If I can get her down to 250 watt-hours/mile, my 22 kw-h pack will get me about 90 miles in a pinch, 75 miles comfortable, and I need 60 miles for my daily commute. I don't expect to have the air conditioning running ever again, but I left it in place in case I change my mind (I know I can save some weight here, but it's easier for non-mechanical people like me to leave it where it is then to keep all of the parts together and install it again)

[thingstodo]

The motor adapter and motor coupler and an emergency disconnect (mounted in an electrical box) were ordered from Canev

Battery pack measurements 24 wide x 13.5 deep x 9.5 high for 16 renault/leaf cells plus the supports on each end. I have 3 of these assemblies, rated 8 kw-h each but they measure out closer to 7.3 kw-h

There is 49 inches across the back seat, seat is 8 inches off the floor plus 3 - 6 inches of foam in the seat, 6 inches in the front sloping down to 3 inches at the back.


Mazda MX6 Day 010 Video 006 July 9 2014 - YouTube

[Astro]

Originally Posted by thingstodo

Wow - you seem to have a strong handle on what everything weighs and what battery capacity you need.

What does you car weigh and what are you projecting for w-h/km or w-h/mile? Does that include some 'fun' driving?

Fun driving has been one of the main requirements from the beginning.

I have a spreadsheet i used to calculate everything right at the beginning of the project. It uses logged GPS data to calculate power usage taking into account such things as the rolling resistance, wind resistance, conversion efficiency, weight, terrain etc.
When i went to get some figures for you i noticed they all seemed a bit off.
I dug down into the formulas and found a fudge figure.
Looks like this figure is used to tune the result to match real world. But that makes it not very accurate as a prediction tool prior to the build.
I didn't notice this in the beginning as i didn't know enough to feel that the figures were not quite right.
So i will have to rewrite the formulas to reduce the need for the fudge figure.
My battery requirements may need some revision.

[thingstodo]

Originally Posted by Astro

I have a spreadsheet i used to calculate everything right at the beginning of the project. It uses logged GPS data to calculate power usage taking into account such things as the rolling resistance, wind resistance, conversion efficiency, weight, terrain etc.

That sounds like a complicated spreadsheet. Perhaps a predictive model would be a better description?

My battery requirements may need some revision.

So are you willing to share? Predicted weight of the car, required battery, depth of discharge, watt-hours required per km or mile?

I'm sure that it's not just me who - others would be fascinated as well!

I'm not asking for the spreadsheet, just some juicy details

[thingstodo]

Put a 12V battery into the car and determine if I removed too much of the wiring harness.

Review of what still works inside the car - dash, windows and locks, heating and cooling, signals, etc

Mazda MX6 Day 014 Video 007 July 13 2014 - YouTube

[Astro]

Originally Posted by thingstodo

That sounds like a complicated spreadsheet. Perhaps a predictive model would be a better description?

So are you willing to share? Predicted weight of the car, required battery, depth of discharge, watt-hours required per km or mile?

I'm sure that it's not just me who - others would be fascinated as well!

I'm not asking for the spreadsheet, just some juicy details

I wouldn't want to post the figures it gave for my car as they were not accurate and may miss inform later readers.

The spreadsheet came with the details for a Mazda MX3 in it.
The accuracy of that existing data is apparently spot on.
The spreadsheet was sent to me by another Ecomodder.
I did a Google search and found it at https://mazdamx3ev.files.wordpress.c...energy-mx3.xls
Much of it is in Slovenian. (Google translate is your friend here)
There is an introduction page in English.
The spreadsheet is fairly straight forward, it's the formulas that are tricky.

It is when i put my data in that the accuracy dropped due to me not knowing in advance what value to use for the fudgefactor variable.

This spreadsheet is crying out to be converted to a web application that would allow people to upload their commute GPS logs and car details and get all the power usage details. (Including regeneration) Maybe even allowing people to share GPS logs so you could see how your vehicle would go in different terrains. (whilst being privacy aware)
Maybe even create some theoretical terrains, like 100klm of perfectly level road to get the theoretical best case figures for your car.

More time, more time, too many projects, not enough time.

[thingstodo]

Driving the transmission with a cordless drill

Mazda MX6 Day 018 Video 008 July 17 2014 - YouTube

- measure the transmission splines
- measure the distance across the engine bay
- assemble a linkage to drive the splined shaft with a drill
- the assembly did not work out - failure on this one

[thingstodo]

Weighing the parts removed from the car

Mazda MX6 Day 021 Video 009 July 20 2014 - YouTube

293 lbs from the front
102 - 122 lbs from the rear
395 - 415 lbs total