Weber and Coulomb's MX-5

[antiscab]

In fact, if we are careless enough to charge one half pack more than the other, then this will indeed happen. It highlights the danger of running paralleled half packs, and we really only intend to run it this way temporarily.

to achieve that extent of current flow you need to have more cells in one battery than the other, not just merely the same number of cells at different SOC

Rob used you add 3 cells to a dump charge pack to fast charge a 45 cell TS battery fairly regularly, to give you an idea of the required voltage difference. even with 3 cells different (and both batteries of same capacity) current was only 0.8C

Matt

[jonescg]

It depends on the cell's capabilities too. I had a 6 volt difference between my LiPo pack and the LiFePO4 cells in the scooter. Hooking it up caused a 50 amp inrush current as they equalised. Enough to get the 8 AWG wires warm.

[weber]

I have a question about clutches. We need a new one because the old one can't take the 350 Nm of our motor. I have found a suitable combination of ceramic/metallic "button" clutch plate and high clamping-force pressure plate. But I need to decide whether to buy the clutch plate with a sprung hub or a solid hub. They are the same price.

   

Tritium James says that from a control point of view, the solid hub would be much better. We are having trouble with jerky (oscillating torque) takeoffs, even when they are quite gentle takeoffs, although this may be related to free play in the drive train instead of, or in addition to, springiness.

But I worry that when changing gears, the lack of springs might cause damaging shock loads on the gearbox input spline. Particularly since the clutch plate is of a type that grabs rather than slips.

Does anyone have any thoughts or experience with this?

[bladecar]

Hi Weber,
Ross Blade, for unknown reasons, changed his point of view about a "sweet spot" to try to maintain while driving.   He decided that one gear only and the revs are ok is the way to go.

The electron stays in 2nd gear at all times. I've only been to a bit over 100 kmh and I don't know the revs though it was whistling a bit

The electron doesn't have a clutch. the Mk5's did gear-changing without the clutch ie very carefully.   Ross is determined that I should not use 1st gear. I have a feeling that he's worried about severe stress on the mechanicals.

Do you need to change gears in general?   Can you do a 2nd or 3rd gear standard?   If so, then I'd be inclined to choose the strongest clutch, probably the solid one. If you had a petrol/diesel car and 'dropped' the clutch when changing gears, you always raised the failure possibility though they're pretty tough. I see it that your aim, with the clutch, is simply to go from one consistend gear to another occasional gear in the most careful way, so that gentle slipping of the clutch in that operation would be used. It just haven't done it myself.

[weber]

Hi bladecar. I may not need to change gears, but I want to change gears, on the fly, quickly, under hard acceleration. It's a sports car.

[Hippie403]

hmmm I wonder how much current an electromagnetic clutch would draw?
(Magnetic powder type)

[BigMouse]

I've opted for a springless hub on my conversion as the springs are not needed. The inertia of the rotor should be less than a crankshaft and connecting rods in an engine, so the torque due to acceleration (deceleration) should not be excessive. Magnetically, an induction motor will never create more than the amount of torque that it's magnetic circuit can support, so if 350Nm is your max, then you shouldn't ever see more than that from your motor under and conditions, accelerating or decelerating.

My biggest concern using any clutch is that most people let off the throttle completely between gears. If the throttle response is fast enough, this could result in a complete stoppage of the motor due to regen kicking in with no load. I'll be programming ramp-rates in to my throttle map to help with that. Basically, if the torque demand goes to zero (rather than negative) for a period of time after letting off the throttle, the motor should respond very quickly to changes in speed during a gear change.

[Johny]

Wouldn't it be cute for the speed/torque control to know when the clutch is in and what gear you are intending to select - then exactly match RPM. You could kind of surmise it from the current RPMs. If high then you are going down a gear, if low then you are going up - just don't skip gears.

[weber]

Thanks Bigmouse, for letting us know you're sucessfully using a clutch with a solid centre. We also have poster zoomzoom on the Aussie MX-5 forum telling us he's been using a solid centre of the specific brand and type I'm considering http://mx5cartalk.com/forum/viewtopic.p ... 57#p761557

And Coulomb reminded me by email that when we go to the higher voltage WaveSculptor we will likely have only 60% of that torque available (but out to higher rpm). Tritium_James has confirmed this. So any additional shock loading due to the energy stored in the rotor and flywheel should still be within the limits of what the gearbox can handle (which the above thread suggests is between 430 Nm and 480 Nm).

Johny, we do have an input to our Driver Controls Unit that tells it when we have either the clutch pedal pressed or we're in neutral. We're not using it yet, but it might be sufficient to have it set the torque to near-zero to avoid the possible regen-stopping problem BigMouse mentions. Although I'm not sure if it can be done fast enough.

[BigMouse]

Thanks Bigmouse, for letting us know you're sucessfully using a clutch with a solid centre.

I've "opted" for a solid disk, but it hasn't even gone in a car yet. That's for my BMW conversion. That parts are in hand, but that car is still driving as an ICE. The clutch in the 300ZX conversion is a 6-puck ceramic clutch, but it has springs, only because I couldn't get one with a solid disk for the same price. The assembly is installed, but it hasn't even been spun up yet.

So time will tell whether it's successful or not, but the reasoning for my decision is as described above.

The regen-stopping issue is one that I expect will come up and need to be accounted for in the firmware. Are there any AC conversions with working clutches out there driving around? They'd be able to say whether or not this is actually an issue.

Edit: Also note that I'm using an aluminium flywheel on both conversions to keep the rotor inertia down. Using a stock steel flywheel might be better from the point of view of stressing the motor shaft and coupling. If you're worried about shock to the transmission, then a lighter flywheel is better obviously.

[weber]

Thanks for the clarification, BigMouse.

This page (and others)
http://www.turbomagazine.com/tech/0407t ... ch_basics/
make it clear that the spline damage that can occur with solid centre clutches is primarily due to the torque pulsations that are an unavoidable result of using an infernal combustion engine -- one per explosion. Not an issue with electric motors.

So I'm going with the solid centre in the hope of minimising control issues caused by drivetrain windup.

[EV2Go]

I doubt you would pick the difference between the two. A button clutch is either in or out. The biggest change in going from organic to metal is trying to learn to take off without slipping the clutch. Since you can start off in gear, the only time the springs would come into play is during gear changes, and even then the difference will be extremely marginal.

On a side note for an electric vehicle I would personally go with as light a flywheel as you can get.When I changed mine in the ICE I shaved 2kgs off the factory setup. Was noticeable at idle made the cams sound even bigger, and it took the motor longer to wind down, the less weight in the ICE made it harder for the engine to decelerate, but in an electric I suspect it would have an opposite effect.

[weber]

Thanks EV2Go. We had to design our own flywheel, with a taperlock hub. You can see the design here: viewtopic.php?t=980&p=25795#p25795 It should have considerably lower moment of inertia than the OEM flywheel as it does away with the 30 mm thick outer ring.

BTW, it seems we do not need any further balancing of our motor or flywheel. I ran the combination up as fast as they would go in neutral with our present voltage for the first time yesterday. It went up to 5800 rpm and I left it there for a minute. It was as smooth as silk. There was not the slightest hint of any vibration at that speed, or on the way up or down.

[Damnthematrix]

Have you onsidered a multiplate clutch? These are often used in high performance racing cars, I even think the V8 fouldoors use them.....

[weber]

Have you onsidered a multiplate clutch? These are often used in high performance racing cars, I even think the V8 fouldoors use them.....

Hi Mike. "NitroDann" on the MX5carTalk forum recommended a dual plate, but they're about $1000 and have way more torque capacity than I need.

I got out of it for AU$492 for a 4-puck ceramic with a rigid hub and a pressure plate with about 50% more clamping force (and hence 50% more pedal effort) than standard, from 949Racing. This price included FedEx getting it to me in 4 days. Thanks to "16bit" on the MX5carTalk forum for pointing me to this one.

Coulomb and I installed it last Friday, as you can see in the following photos. But I have only this evening finished putting everything back together and dropped the MX-5 off its stands.

            

[weber]

There was another job that needed to be done while she was up on stands. This involved the 4 conduits that run up the tunnel where the exhaust pipe used to run. Our certifying engineer recommended that we replace nylon cable-ties with stainless steel so there was no danger of them letting go years hence and wrapping the conduits around the tailshaft.

The 3 orange 25 mm conduits carry hazardous voltage cabling, two of them ("A" and "B") each have two 16 mm^2 single-core double-insulated cables for the A and B half-packs, while the one we call "C" carries the charger cables for both half packs in two 1 mm^2 two-core sheathed cables.

The 10 mm black loom tube carries 8 optic fibres. Two for the A-half-pack battery monitoring, two for the B-half-pack battery monitoring, two for the A charger control and monitoring, and two for the B charger control and monitoring.

These two photos are looking towards the rear.

   

The next two photos are looking towards the front.

   

The remaining nylon cable-ties are not suspending the conduits, but merely bundling them together between the points where they are suspended by stainless cable-ties.

[Damnthematrix]

One question I've been meaning to ask you guys...... rego value is usually calculated on number of cylinders/rotors. HOW did they calculate yours, and how much were you charged?

[weber]

One question I've been meaning to ask you guys...... rego value is usually calculated on number of cylinders/rotors. HOW did they calculate yours, and how much were you charged?

It is treated as having one cylinder/rotor and so 12 months rego was $294.50 including $24.60 for the new number plates. There was no transfer fee or duty because it was previously registered to me. But when it does apply, EVs and hybrids have only 2% duty compared with 3% to 4% for other vehicles. 12 months CTP insurance is the same as for any other car or station-wagon and was $334.60 from RACQ.

Comprehensive insurance for modified vehicles is only available from Shannons, and you must establish that you are a "Motoring Enthusiast". AEVA membership was sufficient for that. They did not want to insure the MX-5 for the replacement cost that I estimated at $55k, but they insured it for $35k with a $500 excess and salvage rights, which suits me just fine.

Salvage rights means that if they declare the vehicle a total loss, I get to keep what's left of the vehicle, _and_ I get the full $35k. For other than a total loss, I can choose my own repairer, including choosing myself.

The first annual premium was $1024.95. If I make no claims (other than a windscreen) the premium is discounted 25% after the first year, 45% after the second year, and after 3 years with no claims a 65% discount is locked in, irrespective of any future claims.

[Damnthematrix]

I was just thinking again about the idea of towing, or in my case, carrying a genset on the back of an EV ute....... and realised it's been done before. It's called a Chevy Volt!

[weber]

I was just thinking again about the idea of towing, or in my case, carrying a genset on the back of an EV ute....... and realised it's been done before. It's called a Chevy Volt!

Er, No. With the Chevy (or Holden) Volt you have no choice but to carry the genset around all the time, whether you need it or not. Oh but wait, because it's always in there, there isn't enough room to fit a decent size battery, so you're going to need the genset more often. Getting a bit off-topic. Try searching on "range extender" in this forum.

[offgridQLD]

Just wanted to say it was great to be able to see this car in person last night. Having the car in front of you and hearing the story of the build process. The good and the bad experiences. It really makes you appreciate what has gone into this car. From a EV perspective it's the most exotic home build I have come across. I'm sure once a they iron out the few bugs and get it together as the intended finished configuration it's going to be a great car -(already is).

Thanks for sharing.

Kurt

[weber]

Just wanted to say it was great to be able to see this car in person last night.

Thanks Kurt. We sure enjoyed finally being able to show it.

BTW, We now know what The MX-5's range is.

It's 15 km less than it needed to be to get my wife and I home last night!

It did 114 km on one charge (27 kWh pack) with about half of that being at over 90 km/h). We managed to crawl off the Centenary Motorway into Sumner Road and stopped right near a 24 hour petrol station. After looking around for a convenient power point we phoned the RACQ. Told the woman on the phone it was electric and just needed a charge. (About 1 hour at our full 20 amps would have let us crawl home). She said they could only tow it, and there was a $60 excess for being more than 10 km from home. We said "What about the roadside generator trailer the RACQ brought to the February Brisbane AEVA meeting?". She put us on hold for some minutes. When she came back she said that her supervisor had not heard of such a thing either.

Here is a photo of the charge trailer, taken at the February meeting.



I remembered the immortal words of Jeff Owen: "There are more publicly accessible power points in Brisbane than there are petrol stations". So I searched further afield. I did find a 10 A power point in an arcade we might have crawled to, but its vehicle gate was locked and so I would have had to park on the footpath and join our two leads together to run them 40 metres, and it would have taken two hours at only 10 amps, and it was raining, and Janelle had to get up early the next day.

We waited about 45 minutes for the tow truck and got home about 11:25 pm.



[Edit: Changed '"It doesn't exist"' to "that her supervisor had not heard of such a thing either". Changed "45 minutes" to "an hour" and "an hour and a half" to "two hours".]

[Adverse Effects]

so are you going to send the RACQ a bill for towing because they lied?

and your evidence is that pic above lol you can even tell them the rego number

[offgridQLD]

Always good to know what your limits are.

Perhaps we need to address the lack of mobile road side charging our self. Even a small Chinese import pure sine generator capable of 15 - 20A could be a very handy asset.

Though it's interesting how time was a factor in your decision to go with the tow. So a generator with high output dose have it's advantages. I have a 8kw 32A kubota Diesel generator that is dedicated to my off grid house. With a over sized PV array and a large battery bank it hasn't been called upon, not even once in over two years.

I was considering mounting it on a trailer so that I could still dock it to the house if need be but take it behind the 4wd if needed. Just to get some use/value from it. Mobile EV rescue service springs to mind.

All said and done I think its much less likely that you will run out of charge again now you know your limits.

Kurt

   

[BigMouse]

Does your BMS report SoC, or do you only know it's empty when it approaches the low-voltage cutout?