Thursday, January 20, 2011

This week we welcome Anthony Bagnulo as a visitor from Ottowa Ontario. Anthony came down for 10 days to work with us and see the EV thing first hand. Apparently, we're not showing the full metal jacket EV building experience in sufficient detail.

In this edition, we have a bit of fun with an oversized plasma ball and digress into plasma ball history and a description of plasma. This will be our indicator to indicate a live cord on the Texaco Fire Chief J1772 charge station.






We also update Speedster Redux which is what we're calling it now. A simple controllerectomy has gone beyond major surgery into a total rebuild. We have a lot of fun stuff going on with it and are really having a ball building this one.

Matt Hauber has just gotten to be quite good at eeking out the final millimeters from a battery box. He worked in two more cells in the front of the car, two more in the rear battery boxes. I added 11 underneath with our "belly pack" which so far has worked out surprisingly well. We have up 1.75 inches in the center of the car from a 9 inch clearance.

Additionally, we decided this week to mount the Soliton1 over the motor, and I must say, it does show off this really quite gorgeous design very well. It also freed up the shelf where we have installed SIX MORE CELLS and a terminal block along with some gland nuts for routing. The result is a very neat installation and a cell total of 57 for a nominal pack voltage of 191 and a pack capacity of 34,380 watt hours. At our previous 225 wH/mile rate, this would indicate a max range of 153 miles.

Unfortunately, we'll be a little over 200 lbs heavier and so roughly 2400 lbs on the road. But it sure has me scratching my head what a 1600 lb version with carbon fiber and aluminum would do.

The center of gravity should be much lower and the polar moment should be somewhat reduced - both good things in Sports cars.

The power available will also be basically TRIPLED. This is almost obscene. The Mini gets up very nicely at 100kw and 3500 lbs. This 170kw estimated at 2400 lbs on Speedster Redux is just overkill - bad design to my way of thinking frankly. But Matt and Brian are talking about taking it to Monacco for the ALT FUEL races in April. I don't know about that.

Implementing J1772 in the charge station is a bit more of a task than in the car. But I like to cheat. I've kind of genned up a circuit I'm toying with. It won't make the 1khz square wave to indicate current available. But it would detect a change from +12v to +6v on the control pilot line as we implemented it in the cars. Understand this would NOT actually comply with SAEJ1772-2010. But it would allow us to hook up to the car with a safely dead line, press the manual switch at the car, and thereby energize the charger without having to walk over to the pump and flip the switch. The plasma ball should be sufficiently visible to tell us we succeeded. And this little circuit could be replaced at any time with a more robust one that actually implements the standard.




If you see any obvious errors, I'd like to know about it before I build and test it. It's a simple comparator. If you press the switch in the car, the 12v through the 1K resistor will drop to about 6 volts from 12v. That will trigger a transistor to turn on a small relay that in turn energizes the two contactors that apply both phases to the J1772 cordset and of course one of the phases to the plasma ball. The manual switch on the side of the Texaco Sky Chief will still work as well.

We're also working on instrumentation - even though this vehicle features the EVISION it had previously. We received our Zeva2 fuel gage driver and added a matching Speedhut fuel gage to the dash. This gage sweeps through about 300 degrees of a full circle - giving us a lot of definition. The second version of this sensor also features a tachometer output for instantaneous current.

It needs some things. It needs a battery backup so you don't lose your AH count when you hit the maintenance switch on your pack. And it needs a reset button so you can synch it up with a full charge once in awhile to reset the cummulative errors that inevitable with this device over time. We're going to do a segment on building a little box with all that in it next edition.

Enjoy.

Jack Rickard

31 comments:

  1. Hey Jack:

    Every once in a while I see y'all doing something I really don't like- like welding without gloves or the original propshaft carrier on the .090 aluminum motor band on the Mini.

    It has happened again. I don't like the idea of structural copper bus bars being used to secure the upper cells in your package tray box of six. There was also a hint in the show that you are using copper bus bar the same way in the belly pack- stucturally, that is, in order to hold the cells in position.

    It seems like a bad idea to retain the cells via the terminal bolts, particularly where these cells are a capital expense and intended to last the life of the car. First, it puts huge transient loads on them, and they weren't designed for that. There's also significant angular and push-pull movement prying against the connection if you are using the terminal bolts to secure the cells. There will further be a fair amount of movement and forces from thermal cycling in the bus bars themselves.

    What's so bad about putting all this force and movement on the terminal connections? First, it puts a lot of mechanical stress into the terminals, anode and cathode, and the cell cases, none of which they were designed as field mounts for the cells. Morever, though, all that stress and movement is going to work those connections pretty hard, and they are very likely to loosen up a little, and produce skyrocketing resistance, especially compared to the other cells in the car that aren't held in place via their terminals.

    Since loose battery connections and high resistance therein are probably the second likeliest way afher a BMS/charging accident to burn one of these cars to the ground, anything that could promote higher resistance in some, but not all of the terminal connections in the car is probably a bad idea. Add in the mechanical stress to the battery terminals and cells themselves from restraining them this way, and I just don't like the practice. If you're doing it in the belly pack, its even more of a problem, because you won't want to check those connections regularly for tightness, nor will you easily be able to inspect them for heat build up.

    Things move around A LOT more in cars than people imagine; especially in convertibles. There was an old railway crossing in Crystal Lake, IL we used to use as kind of a test to see if everything was put together right on our cars. Otherwise it was to be totally avoided! It was 4 sets of C&NW railroad track, at a 30 degree angle to the road, and both the pavement and the wooden ties between the rails (tells you how long ago that was...) were in perpetually terrible shape. It was a horrible stretch of road. Running over that crossing at just 25mph was a true test of a vehicle's structural integrity. A guy I knew cracked the windshield on a Kelmark GT kit car on that crossing. When I chipped the paint on the door jamb of my "solid" 1971 Impala convertible there, which had an assembled door gap of about 3/4", I learned and haven't since forgotten just how much movement there can be in a car.

    Accordingly, holding your batteries in place by the terminals is an idea to reconsider, or at least if you do it watch the terminal connections like a hawk for a long time (maybe always?) until you are sure its really ok. Honestly, though, I just wouldn't do it that way.

    Just a suggestion,

    TomA

    ReplyDelete
  2. Tom:

    You're entirely carried away. Yes, the bus bars and terminal straps brace up and stiffen the pack in all cases. But there is very little force on them - even over the tracks.

    Jack Rickard

    ReplyDelete
  3. Hey Jack:
    Oops, you have an open base on your contactor switch transistor. I know, it would probably work,but it's not good design. Just move the 470 ohm resistor, one diode drop to the east. ;)
    The op-amp has a push-pull output so pull-down isn't needed there.

    Joe Jackman
    Vancouver Canada

    ReplyDelete
  4. Jack,
    You may be interested to know that the control pilot units for the european iec61851 ac charging poles are available from mennekes in germany for 242euros. It uses the exact same voltages and 1khz pilot as j1772. I built one into a little box and it works perfectly with the circuit you published a few weeks back. I actually didnt bother with the switch. Just used a diode and resistor to pull the cp line to 6v. After 2 seconds the contactor pulls in.

    Damien Maguire

    ReplyDelete
  5. Well Jack you went and got the Plasma Ball. Cool. Nice little history. Hope to be seeing your vehicle on the road soon. We also got our Zeva2 fuel gage driver today. That is another move forward with our build.

    ReplyDelete
  6. Damien:

    How about a link partner? I know, my Google must be broken. But a link to a Mennekes page with part number WOULD be nice....

    Jack Rickard

    ReplyDelete
  7. Joe and Jack: Initial I didn't like the open base as well, but since Jack is using an opamp with a worst case output swing 2V -> Vcc - 2V (or so). it is better to use the configuration that Jack has, than to move the 470 ohms resistor to the cathode side of the diode.
    I would have used a single supply rail to rail component or a dedicated comparator like the 339 or similar.
    But then again. This is only a hack for Jack's Fire Chief, and for that it may be sufficient :)

    Martin.

    ReplyDelete
  8. Hi Martin & Jack:
    Martin, you're right. I spent last night, thinking about that and came up with the same 339, but it's a quad chip so I was going to see if I could find a similar single, with the same characteristics.
    Sometimes we old folks (70) yap without enough thinking.
    I still don't like to see a base without some terminating impedance, either + or -.

    Jack, I have been following your efforts for about a year now and all I can say is, if people think you ramble, let them. I don't think you'll lose them. There's too much good stuff here!

    Joe Jackman
    Vancouver, Canada

    ReplyDelete
  9. Sorry! wasn't thinking yesterday. The part number is 30511 for the 32amp version. Spoke with them this morning and its 3 week leadtime as they are still in preproduction. I placed an order for one more for myself and one for you so when it turns up i'll drop it in the mail. I'll email you a couple of photos later today.
    Damien

    ReplyDelete
  10. Ok, ok. I moved the resistor. If even one transistor is saved....

    Jack Rickard

    ReplyDelete
  11. Thank you so much Damien. Very generous.

    Jack Rickard

    ReplyDelete
  12. The LM1458 is a dual in an 8DIP I can still see. It's very inexpensive - about a buck I think. And we're not doing much here - detecting when the voltage on pin 4 goes below 6.5volts. I'll probably use a MOSFET instead of the transistor just for life as it will be on for as long as the car is charging. But you get the idea.

    I installed an inlet in the Mini and charged from the Sky Chief. I'm warming up to the whole concept. The connector is just larger and more solid. Feels good to plug it in. But I have it under the hood on the Mini so I have to open it.

    Jack Rickard

    Jack Rickard

    ReplyDelete
  13. Hello Jack, here's a link to a 1Khz generator: http://cds.linear.com/docs/Design%20Note/dn262f.pdf A 555 based design is also relatively easy.

    JR

    ReplyDelete
  14. Jack, A few comments. I loved the 1/14 show. Don't know why. Maybe because I grew up in Ottawa where Anthony Bagnulo is from. Now live in Toronto ON. I built my own eDart electric Dart. The thing Anthony may not know is there is a controversy right now in Ontario because in the next few years Ontario is going to produce TOO MUCH solar and wind power and there will be a surplus. I kid you not. I believe it is being driven by the non-renewable energy people as the province has entered the renewable energy game in a huge way - they pay anyone to provide power to the grid.

    By the way, you are not being logical about Tesla and battery life/capacity. The analogy is to the Honda S2000. 250 hp but at 7000 rpm -- totally useless for most people but required for marketing. The same goes for Tesla's distance claims. People will buy the cars and only drive 50 miles. They will do this for 8-10 years and never know that the capacity of their batteries has (possibly -maybe not due to lower DOD) declined. Tesla wins the marketing claims AND everyone who buys one is happy with the car because 10 years later it will still do the same 50 miles per charge it always did. Anyway, love the show.

    I also think you need to give Randy at Canadian EV a break about his heaters - he is in BC where the weather is like Seattle's. In Toronto where we have real winters I am installing a Wabasto truck heater.

    Please do a show on the inner workings of controllers.

    Gavin

    http://www.evalbum.com/2379

    ReplyDelete
  15. Love the plasma ball. How about a lava lamp on the next one?

    JRP3

    ReplyDelete
  16. The design of any circuit needs to account for many variable. The one that comes to mind is the effects of the inductor and the collapsing magnetic field (coil) of the main contact. Usually a diode is placed in parallel to counter this induced EMF. However a radial lead aluminum electrolytic capacitor with a operating voltage 0f 50 volts, could be placement at 12 volts and K1 to secure any ripple from the switch.The line side of the supply power should also include Thyristor surge protectors in the event of the electro magnetic contacts fail, lighting or transient power surges. This would behave much like a Ground fault protection circuit without the ground. I think that charging can only be afforded if all precautions are married with an ounce of prevention.

    I can only that all of you for sharing your wisdom.

    ReplyDelete
  17. The only problem with your J1772 setup is that vehicles that are fully compliant won't be able to charge on it. It should work great for your application though since it's a great connector (very, very heavy duty).

    ReplyDelete
  18. Regarding the plasma ball:
    I was wondering, would a rotating magnetic cathode produce a plasma toroid? After thinking about this I believe the magnet does not need to rotate as the field lines of the magnet rotate by themselves.

    Great show as always.

    ReplyDelete
  19. Hi Jack.

    Just watched your new video which I guess you recorded mostly on the 21st Jan.

    You seem to be almost finished with Speedster Redux (or whatever your going to call it), and I assume that either next week or the week after the show will feature a test drive. I'm looking forward to that.

    Can I humbly request that you take out both speedsters on the test drive? My thinking is that if you do that you'll be able to both get some outside shots of the redone car, but also importantly you'll be able to visually show the performance difference between the two which would be very interesting to see.

    I appreciate that doing it the way I'm suggesting would be more of a hassle and take more people, and also that your going to want to be careful with the new car at first for safety reasons.

    Just my two cents worth. :-)

    Nick F.

    ReplyDelete
  20. Hey Jack, watching the Jan 23rd show, really liked the amp counter section in there. Reminded me of The Police's Every Breath You Take video with the one side lightning....

    A couple of comments:

    - Most modern MCU can save variables to flash about 100K times or more over their life so perhaps the Zeva guys can look into a brown out detector and save amp count before loosing power so you can avoid the battery and diode.

    - From the "unnecessarily complex department", perhaps you can hook up your "charged" signal from your charger to a reset circuit on the amp counter so once your charger says your pack is charged, it resets your fuel gauge to full, hands off.

    Regards,
    JR

    ReplyDelete
  21. Top Gear style drag race between the Mini, part duh and the mk II whould be brilliant.
    Padraic

    ReplyDelete
  22. There is a drag strip just south of Cape Girardeau. I didn't know that, but Google did and it told me. 30 miles away.

    Since you live there I guess you knew that. When it gets hotter it could be a fun show if you guys went there.

    Sikeston Drag Strip (389 Sandywood Road, Sikeston, MO).

    Nick F.

    ReplyDelete
  23. Hi Jack
    One thing I've been wondering about for a while (since you mentioned it would have 51, and now 57 battery cells), but still haven't seen you make mention of, is how you're going to do the 50/50 balance-monitoring of your battery pack in Speedster Redux
    Wont the odd number of cells make it rather rather difficult do so?

    ReplyDelete
  24. Why would that be Anthony? It will be 3.3v off. After an initial "eye" calibration, I should know where it is "supposed" to be.

    But yes, if we could work in another cell, all would be well. As it is, we'll still use it. It won't be centered, but it will still be predictive.

    Jack Rickard

    ReplyDelete
  25. I too wondered about the odd number cell imbalance but Lee Hart addresses this in his Batt Bridge circuit by changing resistor values.
    http://www.evdl.org/pages/battbridge.html

    JRP3

    ReplyDelete
  26. The cover Matt is building to go over the six new cells would be a perfect place for an EVTV.me decal

    ReplyDelete
  27. I *think* the battery balance circuit Jack uses has an op-amp that drives a led bargraph display so the magnitude of imbalance can be shown.

    The common point of the circuit is probably the center of the battery pack as well as the reverance voltage to the + input on the op-amp. The positive and and negative sides of the battery pack are then tied to the - input of the op-amp, each through a ~50Kish precision resistor. The - input of the op-amp is then called a summing junction. When the the voltage from either side of the pack is equal but oposite, the current at the summing junction is zero so there is no output. Any imbalance between the two sides of the battery pack causes a current at the summing junction which is countered by a feedback or gain resistor tied between the output of the op-amp and the summing junction.

    Having an odd number of cells in the pack means that getting at half the pack voltage is difficult at best. To center the meter, a couple of things can be done. 1st, a resistor can be added to the side of the pack with the extra cell such that despite it being 3.4v greater than the other side, the current from both sides is the same, resulting in no current at the summing junction. 2nd, a third resistor can be added to the summing junction of the op-amp so as to zero the current at the summing junction. A ~20K pot between the power supply rails for the op-amp with an appropriate resistor from the wiper to the summing junction would allow the user to zero or center the bar graph despite the 3.4v (or more) difference. -Klaus

    ReplyDelete
  28. How about a 3" plasma ball back where the exhaust pipe use to exit.

    ReplyDelete
  29. JRP3: It is a nice simple circuit. But I would do three things before I would use it in my car or any other car for that matter:
    1: Put a fuse on each of the wires from the traction battery - as close to the battery poles as I can get them.
    2: Use three opto-couplers in order to have a galvanic barrier between the traction battery and the rest of the instrumentation.
    3: Put a relay on the circuit so the circuit won't draw any current when the ignition key is in the off position.

    Martin.

    ReplyDelete
  30. Martin:

    1. Use 30 awg wire.
    2. Why?
    3. 20ma is across the entire pack. It does not represent a parasitic load on any specific cell.

    You're overthinking this circuit. It's beauty is it doesn't do much. It just does exactly what you need.

    Jack Rickard

    ReplyDelete
  31. Jack:
    1: I do not like a wire acting as a fuse - in many cases it might burn down the car...
    2: In order not to have lethal voltages running around in the car - better safe than sorry. 300VDC can give a nasty surprise.
    3: It is still 20mA - I do not like the parasitic load and it does not cost me a lot to get rid of it.

    I do agree I might be overdoing it, but where I come from, we do not like to put the customer or the customers equipment at risk.

    I also like fully insulated chargers - remember?

    Martin.

    ReplyDelete