Aptera Features to Support or Avoid

[PatQ562]

OK, folks, my first new thread. Sparked by a report in today's LA Times regarding the problems with keyless ignitions (inconsistent operation, operator confusion, difficulty turning car off in emergencies, and occasional failure to start due to nearby RF interference), now that there is some small hope of a "home stretch", time to start getting our licks in as to how we want the Aptera to actually work.

[PatQ562]

The "ignition key" should work exactly like the standard we have used for the last fifty-plus years.
1. Insert actual key into the lock. (Valet parking NEEDS to have a real key, not an entry code)
2. Turn key to "On". All vehicle accessory systems (lights, radio etc) become active, but vehicle does not yet move. The various readouts are active (state of charge etc).
3. To activate propulsion, turn key to momentary "Start" position. There should be a distinctive "ramp up" sound effect (climbing series of notes, a "whooeee" or some other evocative effect) that tells the driver that power is now available.
4. To drive, put foot on brake, take transmission out of Park, select gear and use the throttle.
5. To kill power, turn key off. Needless to say, this must not engage steering wheel lock, nor should any driving controls become unusable.

AVOID all the uncertainty and needless complexity of the "keyless" systems being foisted on the public. We don't need any more points of failure.

AVOID an entry keypad such as the EV1 used. This was always an issue for valet parking. Aside from not wanting to give away my entry code, valets need a key on a ring to hang at their service desk.

Pat Q

[PatQ562]

The EV1 got it "exactly right" regarding thottle response and "creep". The vehicle behaved exactly like an ordinary automatic transmission. With the vehicle "in gear", it would creep forward or backwards (as selected) when releasing the brake. This allows the driver to "inch" the vehicle up to obstacles by controlling with the brake, rather than switching between brake and thottle, and is especially handy in reverse or when parking.

Needless to say, although it "feels" like the car is slightly straining against the brakes, actual motive power is cut when the brake pedal is depressed.

The maximum creep speed was very modest - 1-2 mph at the most. Any concern about dumping the car "in gear" and having it walk away on you is resolved by enforcing the now-standard interlock that the brake must be depressed before the gear selector can be moved out of Park.

[PatQ562]

Much discussion occurred earlier regarding the ideal amount of off-throttle regeneration. The most intuitive, and logical, means for satisfying various desires is to provide several ranges of "Drive" in the forward gear selector. (Needless to say, regenerative BRAKING is always active when using the brake pedal.)

D1 should enable free coasting on throttle lift-off, with power ramping up from the beginning of throttle travel.
D2 should enable moderate regeneration, where the throttle passes through a neutral zone near zero, below which is a zone of controllable deceleration.
D3 enables strong regeneration, moving the "neutral zone" further into the throttle's range.

It should be apparent that using D2 or D3 will call for more throttle depression for a given speed, since the throttle must first get past the neutral point. Furthermore, the "deceleration zone" will be a dead zone when stopped, that the throttle must pass through before power flows. Both of these factors will feel much like "low gear" in a conventional automatic, and will therefore feel like a natural consequence to drivers.

The degree of regeneration, if any, in Reverse should be a matter for driver testing and controllability.

Overall throttle response on the EV1 was textbook perfect, with smooth off-idle onset, nice incremental response throughout the range, and good pedal feel. The power-vs-throttle response was "torque as a function of throttle position" until the motor reached full voltage, shifting to "power as a function of throttle position" above this. The net effect was a steady rush of acceleration up to about 50% of full speed, then gradually tapering off up to full speed, but always smoothly modulated by throttle position.

The "Gear Selector" itself should fall easily to hand, ideally in the normal position on the center console, so it is easy to change regeneration at will, or rock the car between forward and reverse to get unstuck.

Pat Q

[PatQ562]

The EV1 devoted a great deal of effort towards masking the difference between regenerative and mechanical braking. The system involved chronic clicking of relays every time the pedal was touched, with an elaborate system to create artificial "feel" of brake resistance during regeneration. Most of this was unnecessary and annoying.

The Honda Civic Hybrid resorts to the obvious method of using the normal pedal travel that occurs before engaging the mechanical brakes for controlled regeneration. Drivers quickly learn that "feathering" the brakes in this zone is an easy and efficient means to reduce speed, with hard braking always available when needed. It's easy to learn and makes you feel smart, but a first-time driver can just step on the brakes as always with normal results. This should be entirely adequate for our purposes.

[PatQ562]

One of the biggest "unforced errors" in the EV1 design was the failure to enable at least a quick shot of INSTANT HEAT on demand. The heat pump (AC unit in reverse) was very slow to warm up, and never came on at all on moderately cold nights. This was to improve range, of course, but a couple of minutes of instant electric heat, and probably, use of heated seats, would make the driver feel cozy without unduly draining the batteries.

[PatQ562]

It's probably too long a topic to march through every light, switch, knob etc but Honda is highly regarded for "getting it right" with simple, visually clear, and nice-feeling controls. Hopefully we will avoid the nightmare of traditional GM controls that always felt like they were falling apart in your hands when used. It WOULD be nice if the radio and other controls were within easy reach without having to lean forward.

[NeilBlanchard]

Hi Pat,

You have my vote on all your points!

Would heated seats be a more efficient way to warm the people in the vehicle? The passenger's seat would would only be heated if someone was in it...

[Schrodinger]

The "ignition key" should work exactly like the standard we have used for the last fifty-plus years...


I'd have to respectfully disagree with you there. As someone who drove a Prius for several years, I must say that the keyless entry and ignition was a welcome convenience. No more lock-outs, as you can't lock the car from outside unless there is a key outside with you. You can leave the key buried in a backpack, purse or pocket and not bother fishing it out. No more worrying about the other two pounds of keys on the ring (or carabiner) hanging from the ignition and damaging the mechanism.

Valet compatibility is a non-issue as you can just hand over the key fob. There was also the option for a manual override for both vehicle ingress and activation, which takes care of your high RF interference issue.

The only thing I would change about the system would be to put a small buzzer in each fob that gave off an audible signal in response to some RF call. That way I can use one key to find the other buried under a pile of laundry. ;)

[evmavin]

The D1-D3 modes should not impact regen as it restricts options. If people want to "dial in " regen there should be a control that adjusts it form 0 to full on any mode so the user can choose. The D modes are there for people who don't know how to drive an EV efficiently under varying conditions and to keep the consumption down. What if you want performance and no regen or performance and full regen or performance and med regen? If you lock regen to modes then it may not match driving style or needs but if it is adjustable you can. I don't need an economy mode to achieve efficiency but most drivers do and I do like the ability to have aggressive regen at times with performance so that would appear to be a contradiction in terms.

Switches on a brake pedal for regen are a bit basic since you can use a HEPI control and get linear feedback if desired but I would guess they use a switch. I would guess they use one level on the brake and one with off accelerator and that may be it, I'll just put an adjustment on the off pedal and be done with it. Regen has come a long way in modern AC EVs. You really did not want to open this can of worms since most people have never driven a car with adjustable regen and don't understand the value and benefit in an (EV). The majority of drivers just need brake regen that is similar to regular braking with low regen off pedal, the difference in efficiency for most is nominal. If you drive aggressively and stop short you will want high off pedal regen. For most KISS will work.


As far as the ignition goes, sure a key will work just like a normal ignition. Both my EV's worked this way and there is no "delay" or "startup" it's just faster than a normal car to start, just a click of a contactor and a "on'" indicator. If they can find a reliable RF system I'm for it since it avoids key holes and is much more in making with the design, RF issues are many times the result of poor engineering. I work with RF all the time and the issues with interference can be mitigated with good receivers. I really don't want to see a lock on the vehicle, it already looks too much like a car with all the changes and who is really going to valet park an Aptera? Do you really want the front end torn off and could you imagine explaining it to the valet? Perhaps in two years that will be easier.


As far as ergonomics, the old dash was a very nice design, the new EVA renderings look like, well crapy bling. If it's very ergo and looks like a plastic toy people will be put off. The interior seems to be looking far too much like a GM product with much gloss, as long as it's functional and looks clean I'm ok with it but the dash seems to look like some overdone robot.

[evmavin]

Hi Pat,

You have my vote on all your points!

Would heated seats be a more efficient way to warm the people in the vehicle? The passenger's seat would would only be heated if someone was in it...


The car is so small and insulated that heated seats are a waste of energy. What happened to the heat pump?

[NeilBlanchard]

Read through Pat's posts -- his experience with the EV1 and it's heatpump were not so wonderful...

[aptera1213]

There was also the option for a manual override for both vehicle ingress and activation, which takes care of your high RF interference issue.

So they should at least have a manual override I hope. I know that we have huge RF interference at the top of the mountain near where I live (lots of TV and radio towers). My scooter has remote unlocking of the storage compartment from my key and it won't work on the mountain top. Luckily the scooter starts from the key itself and not the remote or I would have been stuck up at the top of the mountain a few times.

Gavin

[evmavin]

So they should at least have a manual override I hope. I know that we have huge RF interference at the top of the mountain near where I live (lots of TV and radio towers). My scooter has remote unlocking of the storage compartment from my key and it won't work on the mountain top. Luckily the scooter starts from the key itself and not the remote or I would have been stuck up at the top of the mountain a few times.

Gavin


If you are speaking of Sutro tower by chance then that would be a huge exception since nothing works near there but it is an unusual situation.

[lapwing]

Bugger! You are a conservative lot!

You want to drive a vehicle as radical as the Aptera, but it should operate like a ....... (name your current ride)!

I am disappointed!

The best driver interfaces I have come to grips with so far. The Prius, and that can leave a lot to be desired!

Key-less entry with RF/capacitance touch. Best thing about the Prius. Fob stays on my person. I would prefer it if my smart cellphone and an Aptera applet could securely interact to relieve me of the fob.

Missing on the Prius. The big red "ooh crap" button, right out in front. This is a low energy, dedicated "contractor cutting the battery pack button" Every electric car should have one. In that circuit a few inertia switches etc. for cutting the power. Air bag deployment should drop that contractor also.

The big fuggly red button should be there.

Turning on the car. No keys, no fobs. It should be enough that I sit down in the seat, put my foot on the brake, a hand on the wheel and and select forward or reverse.
No button pushing, no key turning, the car should acknowledge me personally, not in some futuristic voice, but by adjusting my preferences as I sit down.

These might include climate control, audio, lighting, and everything else adjustable. Heck, I don't even think a steering wheel is the way to go. I would prefer a joystick and some comfortable armrests. Fly by wire! Throw out the steering column. LOL. Who needs the weight.

Yep lets fly this thing! I don't need no stinking wheel! Just the brakes, the joystick and the big fugly red button! ;)

[evmavin]

Bugger! You are a conservative lot!

You want to drive a vehicle as radical as the Aptera, but it should operate like a ....... (name your current ride)!

I am disappointed!

The best driver interfaces I have come to grips with so far. The Prius, and that can leave a lot to be desired!

Key-less entry with RF/capacitance touch. Best thing about the Prius. Fob stays on my person. I would prefer it if my smart cellphone and an Aptera applet could securely interact to relieve me of the fob.

Missing on the Prius. The big red "ooh crap" button, right out in front. This is a low energy, dedicated "contractor cutting the battery pack button" Every electric car should have one. In that circuit a few inertia switches etc. for cutting the power. Air bag deployment should drop that contractor also.

The big fuggly red button should be there.

Turning on the car. No keys, no fobs. It should be enough that I sit down in the seat, put my foot on the brake, a hand on the wheel and and select forward or reverse.
No button pushing, no key turning, the car should acknowledge me personally, not in some futuristic voice, but by adjusting my preferences as I sit down.

These might include climate control, audio, lighting, and everything else adjustable. Heck, I don't even think a steering wheel is the way to go. I would prefer a joystick and some comfortable armrests. Fly by wire! Throw out the steering column. LOL. Who needs the weight.

Yep lets fly this thing! I don't need no stinking wheel! Just the brakes, the joystick and the big fugly red button! ;)

Not sure why you need a button on the Prius as it has failsafe devices which cut the power and if you want to cut the power manually via a button it won's fit on the dash, sorry.

[evmavin]

Read through Pat's posts -- his experience with the EV1 and it's heatpump were not so wonderful...

Good thing they are not using old GM EV1 parts. That's like the argument of power steering being inefficient and heavy but is is not and it make the car drive better but most people have no practical experience with many of the new technologies available today. They may not use a heat pump and if not it will be a basic heater coil but then where do you get the AC.

If we really want to be old fashioned and stick to keys in the doors and ignition on such a modern car we can use a volt and amp meter to determine range as it is more reliable than a computer. Time to more forward as there are reliable systems available today and this in not a model T.

[Apt3448]

Good thing they are not using old GM EV1 parts. (...) Time to more forward as there are reliable systems available today and this in not a model T.

Gotta love progress. I remember clearly the excitement of the EV1 - look, no tail pipe! And now, a few iPods later, you cal it a model T. Doesn't make me feel any younger....

[evmavin]

Gotta love progress. I remember clearly the excitement of the EV1 - look, no tail pipe! And now, a few iPods later, you cal it a model T. Doesn't make me feel any younger....


The EV1 is by no means, the point is the 2e should not be made into a low tech car for fear of electronics. Key fobs are the least challenging items in the Aptera challenges.

[lapwing]

Not sure why you need a button on the Prius as it has failsafe devices which cut the power and if you want to cut the power manually via a button it won's fit on the dash, sorry.

If you own a Gen 2 Prius then you got the recall notices regarding accelerator peddle issues with the floor mats.

A number of drivers got into trouble, with stuck throttle peddle and no clear way to switch off the vehicle. (actually shifting to N or punching the P park button, would disengage warp drive and of course pressing the power button would work. However, in a panic an unambiguous, huge red fuggly emergency button is what is needed to trigger a valid FAP (fixed action pattern), and save a nasty stuck throttle situation.http://pressroom.toyota.com/pr/tms/toyota/toyota-consumer-safety-advisory-102572.aspx

[evmavin]

If you own a Gen 2 Prius then you got the recall notices regarding accelerator peddle issues with the floor mats.

A number of drivers got into trouble, with stuck throttle peddle and no clear way to switch off the vehicle. (actually shifting to N or punching the P park button, would disengage warp drive and of course pressing the power button would work. However, in a panic an unambiguous, huge red fuggly emergency button is what is needed to trigger a valid FAP (fixed action pattern), and save a nasty stuck throttle situation.


Thats just bad design on toyotas part and that can happen in any car, the start button should reduce the motor power if pushed, so should the brake. On my EVs it the brake is pushed while accelerating the motor powers back. The cause was a defective mat design.

[OC-LA driver]

I love the Prius' keyless entry, but I also like having a power button to turn on/off not just have it fire up every time I sit in it to find my sunglasses.

I don't think the big red button is needed. I once suffered complete brake failure on the freeway, but was able to use light throttle and downshifting (automatic transmission) to slow enough to get off the freeway and use emergency brake to stop...even though my mental function was, let's just say, a little impaired that evening (30 years ago). With a big red button I might have just stabbed it and then been locked up and skidding, or perhaps in neutral, coasting with no control at all. If a stuck accelerator is overwhelming the brakes, neutral works fine, and gives the driver a moment to think, possibly fix the problem, and decide whether to re-engage the engine.

[wpatters]

I think we gave up the early access to the Aptera when the engineers didn't stop modifications.

If we still had a belt drive primitive motorcycle-like 3 wheeler, we would be driving it now.

[lapwing]

.............. The cause was a defective mat design.

That was one problem, but faulty pedals may be an issue also.

[RainCaster]

I really enjoy the keyless ignition/locks on the Gen 2 & 3 Prius and Lexus sedans. Toyota has done a fine job with these, and I have never seen any of the issues mentioned in that article. This is a feature that I would like to have.

Also, I want bluetooth phone support for the stereo, along with USB and AUX inputs.

[lapwing]

I love the Prius' keyless entry, but I also like having a power button to turn on/off not just have it fire up every time I sit in it to find my sunglasses.

................

Agreed, sitting down should only activate the courtesy mode. Interior lights, audio and cellphone. T

To boot the car to main battery pack and EMU, takes the fob on your person, foot on the brake and hand to the steering (or joystick :happy0017: ) And big red button in the out position.

The park interlock could be integrated into the big red button, thus establishing a FAP. In the Prius pressing the park button while driving engages neutral, but does not engage the park transmission interlock, until the vehicle is stopped.

On the Aptera I can see me pull up to park, hit the big red button and the EMU disengages the main pack, and if there is no wheel motion engages the park transmission interlock. After say 5 minutes the computer shuts down - no further button pushing required. If there is wheel motion the MCU remains on but in neutral.

[Rat]

The car is so small and insulated that heated seats are a waste of energy. What happened to the heat pump?
There was just an item on AutoblogGreen that the Volvo C30 EV due out eventually (2013?) uses an ethanol heater for cabin temperature. That avoids draining the battery power, which is especially a problem in extreme cold. It's cold there in Sweden. I thought this was clever. It makes the car sort of a hybrid since it uses both electric power and liquid fuel for its energy needs, but all the mechanical and propulsion of course uses just the battery power. I admit I'm a weather wimp, which is one reason I live in California. I need a well-heated vehicle in cold weather. I don't need a seat heater, although my Volvo has one, but it's important for me to know I can turn on the heater without drastically reducing my range.

[KarenRei]

What might be nice would be to pair it with a thermoelectric generator to help run the accessories (say the fan). Adds almost no complexity (unlike a full generator), and while the efficiency is low, some power is certainly better than none. I've never looked into the costs, though.

[PatQ562]

Although many respondents provided input requesting various modern convenience features, such as keyless ignition, and of course, if that's already designed in, I wouldn't want to delay things by changing it, you guys are dreamin' if you think a small start-up can duplicate the work of huge conglomerates without introducing many new failure modes! The whole point of my rant against keyless ignition was triggered by newspaper reports about user problems despite billions of dollars of development. The several examples listing RF "problem areas" only proves the point. Once is enough to be a non-starter.

I do however believe Aptera had plans to support a fully modern entertainment and navigation system, which I have no inherent objection to since it's not essential to getting home in the dark.

Pat Q

[PatQ562]

Whether or not the "standard" keyed ignition is accepted, there should be a way to tell at a glance whether the "motor" is on. This can be supported by enabling the MOVING display, such as the speedo, only when power is enabled. So turning the key on would light up the status gauges such as charge, but "Start" is necessary to light up the speedometer digits.

All this said, I have no particular objection to a Start button instead of the extra key turn, but it seems less convenient.

In a perfect world, I would love it if my house, office doors, car, ATM all "knew who I was" and provided service on sight like a faithful retainer. I may live to see such technology become cheap and reliable, but it's a good generation away and I may not make it. I'm not counting on Aptera to have this fully developed right now.

Pat Q

[steve]

not a reason for delay. We decided to move to FWD in Jan-2008. The first FWD prototype drove in March 2008.

[evmavin]

Although many respondents provided input requesting various modern convenience features, such as keyless ignition, and of course, if that's already designed in, I wouldn't want to delay things by changing it, you guys are dreamin' if you think a small start-up can duplicate the work of huge conglomerates without introducing many new failure modes! The whole point of my rant against keyless ignition was triggered by newspaper reports about user problems despite billions of dollars of development. The several examples listing RF "problem areas" only proves the point. Once is enough to be a non-starter.

I do however believe Aptera had plans to support a fully modern entertainment and navigation system, which I have no inherent objection to since it's not essential to getting home in the dark.

Pat Q



I get your point but when it comes to certain devices like OEM car alarms and radios some were horrible, the RF issues were terrible and yet you could buy third party products that worked perfectly. Conversely, I would not want Aptera building inverters, motors, brakes, etc. This really depends on the device and who makes it and how much use it gets in other markets. Some parts OEM car makers do well and some they don't and some also buy "cheaper" parts from 2nd tier suppliers to save money when there are more reliable options available. I can think of many examples of electronics that don't work in some cars but well in others as a result of supplier issues. In fact GM started getting their "factory" car alarms from a known aftermarket company because the ones they had made for them initially worked poorly. I know of third party RF units that work with 50% success in even good conditions and ones that work every time. The examples are endless because auto makers were never good at electronic features and some sourced the right stuff that worked.

[evmavin]

not a reason for delay. We decided to move to FWD in Jan-2008. The first FWD prototype drove in March 2008.


Yes, exactly, and a good decision it was:)

[evmavin]

Hypothetically speaking, if Aptera were to get no DOE money and close how is the intellectual property structured and could someone restart the company fresh and who would need to be paid for the rights in basic terms? Would the IP be held hostage for a large some or is there are possibility to buy it at a low price after closing and restart the company at an advantage of cleaning out the "low value" overhead. If this were true I could most likely get adequate funding to restart such a venture and would be willing to do so. My concern is if they get a large loan that it will be managed poorly, personally I would invest my own money and work at a salary that would coverless than my basic living costs. If people like MM are getting paid more than $40K I would be disappointed from many fronts, but that's another rant I won't start.

[PatQ562]

Front wheel drive is a single mechanical system that you can practically buy off-the-shelf, at least in prototype form (Azure Dynamics has a motor/gear/diff set all ready to connect half-shafts up). Believe me, the challenges of designing, debugging, and programming the latest generation of wireless key-recognition systems, let alone fully modern vehicle dynamics control and interior automation, are MUCH MUCH harder than converting to FWD. Look at all the questions we've raised here about what is even wanted, before design can even begin, and we've barely scratched the surface. FWD is a "no brainer" by comparison.

In my opinion, a 3-wheel vehicle like Aptera will pose enough handling and reliability concerns to justify utterly conventional, no-brainer controls wherever possible. Save the fancy doo-dads for Gen 2.

That said, I would value electric locks (with manual override), and variable-speed wipers, but I still like the security of sticking the ol' key in the ignition.

This brings up the whole subject of "what makes a basic car charming?" In my opinion, a true economy car "makes do with less", but is arranged in such a way that you can enjoy a pleasant driving experience with a few basic tricks that make you feel smart. Most modern cars (let alone computers) have way too many features that ultimately make you feel stupid for not knowing how to use the machine. I sense that many (but not all) Aptera prospects want an appealing economy car, even though the early-adopter price is at a premium. What used to be a pure appeal to the pocketbook (as with the VW bug) has become an appeal to reducing impacts on the planet, as well as the pocketbook. I personally am no longer charmed by the modern over-elaborate vehicle, with its layers of driver isolation, safety systems, electronic distractions, and novelty based "conveniences" that require special learning and generate ongoing frustrations (how many times have I struggled to open my car's back door in the rain, because I keep forgetting that to unlock ALL the doors, I have to click Unlock TWICE? - and I HATE those cars that lock you in automatically, leaving me tugging on the handle when I want to get out). As a driver of vintage cars, I appreciate the honesty of a simple machine whose function sits right where you can see it - but I will admit old cars are primitive in terms of noise as well as pollution. An electric car offers a chance for a nice civilized ride without the need to bury everything under layers of isolation - and if the weight can be kept down, we can omit the cost and long-term unreliability of power steering, power brakes and so forth. A small 2-door car where you can reach over and lock the passenger door before getting out practically eliminates the need for electric locks. I'd much rather move a lever and cable to open a ventilation flap, than flip a switch that HOPEFULLY opens a flap somewhere if it's still working. At least with the lever, you can FEEL if something's stuck. And I'd much rather move an AC control to "hotter/colder" than wonder what exact temperature I want to program into the "climate control".

Modern world-class producers (mainly Japanese) have performed wonders in delivering incredibly complex cars that really do work flawlessly 99.9% of the time (Toyota's throttle problems show that even the best don't walk on water however). BUT Japanese cars are designed for a 6-year service life and aren't supported indefinitely. Those of you who plan to keep their Apteras indefinitely had better think about long-term maintenance and what you're gonna do when the wireless key mysteriously stops working, and the seat-weight and "hand-on-wheel" detectors fail.

Pat Q

[evmavin]

My father was an aerospace engineer. I respect the ignition key bit but this is an Aptera and engineers are too practical at times and I'm glad my dad was very eccentric. My Halloween costume in the 70's had B1B technology so toggle switches were not very exciting to me and although I may may have had STP stickers on my Schwinn, I was thinking EV's when I drove 12V into a sizzler motor and made it beg for mercy until I could smell the brushes melt. Purple Otter Pops and Fun Dip with all their synthetic coloring should not stand in the way of forward thinking, not even those Astronaut stick things. Put those slide rules aside and think post modern with a double twist of modern, this is what brought this vehicle here in the first place.

[Rat]

My father was an aerospace engineer. I respect the ignition key bit but this is an Aptera and engineers are too practical at times and I'm glad my dad was very eccentric. My Halloween costume in the 70's had B1B technology so toggle switches were not very exciting to me and although I may may have had STP stickers on my Schwinn, I was thinking EV's when I drove 12V into a sizzler motor and made it beg for mercy until I could smell the brushes melt. Purple Otter Pops and Fun Dip with all their synthetic coloring should not stand in the way of forward thinking, not even those Astronaut stick things. Put those slide rules aside and think post modern with a double twist of modern, this is what brought this vehicle here in the first place.
Maybe I'm an ignoramus, but I don't understand this, at least not after "My father was an aerospace engineer." Purple Otter Pop? Fun Dip? Sizzler motor? [I thought they sold minuscule gristle steaks with garlic toast, not motors.] It's a fun read, though.

[evmavin]

Maybe I'm an ignoramus, but I don't understand this, at least not after "My father was an aerospace engineer." Purple Otter Pop? Fun Dip? Sizzler motor? [I thought they sold minuscule gristle steaks with garlic toast, not motors.] It's a fun read, though.


Someone here wil get it, It's a 70's thing:)

[KarenRei]

Sizzler motor: wasn't that from slotcars?

Otter pops:

http://www.byrnerobotics.com/forum/uploads/BillMimbu/2009-09-17_191645_Otter-Pops.jpg

Fun dip:

http://newscoma.com/wp-content/uploads/2009/06/290px-Fundip.jpg

Remember the Ring Pop? Or was that after your time? :)

[evmavin]

[QUOTE=KarenRei]Sizzler motor: wasn't that from slotcars?

Exactly, and they were not 12V.

[lapwing]

One feature I am sure they have not got right, the "State of charge" and it's interface.

This single most disappointing feature of every EV.

Most every "gas-gage" needs a full charge to reset it. After several partial charges/ discharges, the ones I've come across are not even close to reality. Throw in a bit of cold weather, and stuck on the side of the road comes to mind. http://green.autoblog.com/2010/01/27/determining-state-of-charge-of-lithium-batteries-problematic-for/

[KarenRei]

You're referring to the Mini-E, I take it?

That was a hastily done conversion car with all sorts of bugs that they charged the owners a mint for the privilege to drive. And yet the lessees still loved it. The key is that they got them out there to those who were willing to be early adopters so that they were getting important real-world data on what worked and what didn't, and then used the knowledge to fix problems on the early adopters' vehicles and those of all future lessees, as well as to further their EV program's development as a whole. If they had been a startup, the fact that they had been actually producing and selling vehicles would have been a tremendous boost in their chances of landing funding.

Hmm, I wonder if there's a lesson in there somewhere.... ;)

And that's not even really a fair comparison... to Aptera. What they were working on in 2008 was on track to be far higher quality than what BMW put out with the Mini-E program. But a certain CEO kept refusing to give the go-ahead to actually commit to the tooling for almost any part and kept ordering redesigns, something that continued all throughout 2009.

[NmGfan]

One feature I am sure they have not got right, the "State of charge" and it's interface.

This single most disappointing feature of every EV.

Most every "gas-gage" needs a full charge to reset it. After several partial charges/ discharges, the ones I've come across are not even close to reality. Throw in a bit of cold weather, and stuck on the side of the road comes to mind.

I have to disagree on that one. After using the e-meter in my NmG for over two years, I find it to be very accurate. I've learned what parameters are important (voltage level) and which to blow-off (percent full). The Bay Area has mild temperatures, so winter and summer range around here (on Lead Acid batteries) doesn't vary that much. Interestingly, I find most errors in the state of charge to generally be in my favor, meaning there is more power available (as much as 500W), not less. Partial charges are actually pretty accurate and have not caused any errors that result in less range than indicated on the e-meter (baring that heavy right foot).
:happy0025:

[evmavin]

I have to disagree on that one. After using the e-meter in my NmG for over two years, I find it to be very accurate. I've learned what parameters are important (voltage level) and which to blow-off (percent full). The Bay Area has mild temperatures, so winter and summer range around here (on Lead Acid batteries) doesn't vary that much. Interestingly, I find most errors in the state of charge to generally be in my favor, meaning there is more power available (as much as 500W), not less. Partial charges are actually pretty accurate and have not caused any errors that result in less range than indicated on the e-meter (baring that heavy right foot).
:happy0025:


This is because the e meter is good at measuring what goes in and out via a shunt and measures the current passing through the pack. Once calibrated on a lead pack they are quite accurate and they do "reset" on a full charge. On the NMG you are better changing the default e meter settings to read in KWH rather than Ah as it is a more accurate representation of actual use.

[lapwing]

I have to disagree on that one. After using the e-meter in my NmG for over two years, I find it to be very accurate. I've learned what parameters are important (voltage level) and which to blow-off (percent full). The Bay Area has mild temperatures, so winter and summer range around here (on Lead Acid batteries) doesn't vary that much. Interestingly, I find most errors in the state of charge to generally be in my favor, meaning there is more power available (as much as 500W), not less. Partial charges are actually pretty accurate and have not caused any errors that result in less range than indicated on the e-meter (baring that heavy right foot).
:happy0025:

Lead acid are quite nice that way. Much harder to do with lithium where voltages actually are not very useful for predicting state of charge.

[NmGfan]

This is because the e meter is good at measuring what goes in and out via a shunt and measures the current passing through the pack. Once calibrated on a lead pack they are quite accurate and they do "reset" on a full charge. On the NMG you are better changing the default e meter settings to read in KWH rather than Ah as it is a more accurate representation of actual use.

Absolutely!! I figured that one out within the first month of ownership. It seemed completely useless set for Ah, so I switched to kWh. I actually pay close attention to the voltage level and kWh consumed (above 139V and about 2kWh for my round trip commute).
I also have installed a PakTrakr that I primarily use to monitor the individual battery voltage levels and SOC and can now(!) collect performance data. It also gives a double check on Pack voltage, current draw, kWh consumed, battery condition, etc.

:happy0025:

[KarenRei]

Lead acid are quite nice that way. Much harder to do with lithium where voltages actually are not very useful for predicting state of charge.

That's what coulomb counting is for ;)

I have to wonder if the Mini-E is just looking at voltage. That could explain the problems.

[evmavin]

That's what coulomb counting is for ;)

I have to wonder if the Mini-E is just looking at voltage. That could explain the problems.

Somehow I doubt that.

[KarenRei]

I am curious as to what your theory is to why their estimation could be so off. That's not typical for li-ion batteries with charge based on coulomb counting.

[evmavin]

I don't know, I think the mini e was rushed and the info and feedback can be taken with a grain of salt. My point it using voltage alone is useless as a measure. I can find out if you really want to know.

[KarenRei]

I don't know, I think the mini e was rushed and the info and feedback can be taken with a grain of salt. My point it using voltage alone is useless as a measure. I can find out if you really want to know.

It's just a curiosity, but if it's not too much effort, I would like to know. And I agree that it's useless as a measure.... which could explain why their charge gauge is kind of useless ;)

[lapwing]

That's what coulomb counting is for ;)

I have to wonder if the Mini-E is just looking at voltage. That could explain the problems.

The problem with just coulomb counting is drift. It's a chemical thing.

Bottom line is there are always more coulombs of electrons in than out, and even when just considering the out, how much you get, changes depending on a whole list of parameters.

This is less of a big deal if every time you charge, you hit the the upper stops and reset the count. Trouble is avoiding drift from reality after a 1/2 dozen partial charges.

So what parameters affect coulomb counting. Battery temperature, battery impedance, discharge rate, regeneration - reversing current flow repeatedly at highly variable levels, as the roadway undulates, or traffic dictates acceleration or deceleration ....... battery capacity and impedance mismatch - I am sure you can add a few more.

Bottom line is, in terms of coulombs what goes in, is different from what goes out and measuring and modeling and taking into account even the major players affecting that is a tough job. Do-able but tough!

I did some calculation this morning, and figured current and voltage counting needs to be really accurate, perhaps 1/10th of one percent measured, or better. Some kind of fuzzy logic model of the main factors like temperature, and charge/discharge rate, need to be integrated into a model (maybe just accurately measured tabulated values), but "correcting" the coulomb counting on the fly.

Not trivial. The awesome very flat voltage v/s state of dis-charge that makes LiFePO4 such a wonderful performer, is also what makes it so difficult to keep track of or determine the true state of charge of a pack on the fly, especially after a bunch of part cycles & part charges.

It's the biggest technical challenge left :character0009:

[lapwing]

It's just a curiosity, but if it's not too much effort, I would like to know. And I agree that it's useless as a measure.... which could explain why their charge gauge is kind of useless ;)

Simple example of how voltage alone lies.

At 20 degrees C hard acceleration pulls a nearly full battery down from 3.2V per cell to 2.9V per cell, as acceleration eases up cell voltage recovers fast to 3.1V The battery was never near empty on a Volts based "gas-gage" that is doing some 3 or 5 minute averaging.

Lets say its 0 degrees centigrade out now. That same almost full pack under acceleration with the same current, has a hard & fast volts dip, into the low 2V range, maybe even 2.0V where the low voltage cutoff wants to "protect" what is an almost full battery. (the cells are in no danger of damage) A meter based on voltage alone even 5 minute averaging is going to report a near empty battery pack at these temperatures even as you ease up on the throttle, as the voltage will stay lower, longer, and lower than at more moderate temperatures. Any software protecting the batteries will get it wrong, well before there is any real danger of over discharge.

Similarly if it's 45 degrees C out and cooking then it would be way to easy to trash the battery using just volts to determine state of charge.

My guess is that the mini uses a temperature sensor and voltage, to approximate a state of charge, and under ideal conditions with large battery packs that are used at very low C-rates, temperature compensated voltages sort of work. However, as soon as you introduce some serious regen into the mix, things get a little iffy. Impedances change depending on rate of charge/discharge and temperature. The volts move up and down a lot, making usefulness as a SOC indicator even with weighted averaging and other tricks pretty much a waste of time.

It's a chemistry thing that I am a long ways from understanding, but I know it can be measured and modeled empirically, and probably mathematically. :happy0005: I will bow out at this point. Others may be able to add more?

[KarenRei]

Indeed, I'm well familiar with li-ion discharge curves at different temperatures. I've had to model them. ;)

That's an interesting theory on what Mini is doing. I am quite curious as to what the real answer is.

[evmavin]

Indeed, I'm well familiar with li-ion discharge curves at different temperatures. I've had to model them. ;)

That's an interesting theory on what Mini is doing. I am quite curious as to what the real answer is.


There is a rumor on ABG that they are using a mechanical dip stick in combination with a shunt and temp sensor that uses x-ray averaging. I'm not sure about the dip stick technology but the temp sensor aspect seems solid. I will call Tom and find out.

[PatQ562]

All this discussion about state-of-charge reveals a valid, and tricky, problem with EV BMS systems. The EV1 SOC and range meters were pretty good, but despite occasional automatic "topping charges", I did experience increasing drift and unexpected shortfalls especially towards the end as the battery pack started weakening. This is why I "hope" Aptera is conducting extensive charge/discharge testing at various temperatures to capture cell performance metrics, to supplement and check on straightforward coulomb counting. There IS a need to capture the end points from time to time. The low self-discharge of lithium should help with the other bugaboo, the gradual development of cell imbalance that causes unexpected loss of power when the weakest cell poops out even though the overall pack is still at at fair voltage. Advanced BMS systems can help with this.

I made my own coulomb counter for the electric Karmann-Ghia, mainly to measure charge/discharge efficiency, and it was useful, but not as valid as watching voltage sag-under-load, and simply getting some history on what to expect. One finding, typical of all lead-acid batteries I've tested, is that less than half the nominal Ah capacity was actually available at realistic EV discharge rates.

Pat Q