Not as serious as a reprap tantrum, my RS485 bus is not behaving very well. If I terminate it properly, it doesn't work at all, and if I don't terminate it at all, then it works, but charging at more than 10A causes a lot of extra characters to appear. Termination has two functions, first, it absorbs the energy in each character as it hits the end of the bus. Without terminators the pulses bounce off the end of the bus and travel back the way they came, causing interference. The second function of the terminators is to hold the bus in a relatively low impedance state, so any stray energy (say that created by the electric and magnetic fields generated by the charger) that gets onto the bus is absorbed without causing spurious characters.

Without terminators, I'm seeing extra characters, but with a 2 metre bus, I don't have problems with reflections.

So why doesn't it work with terminators? It turns out the transmit enable circuit cannot predict the future. With no cells transmitting, the bus floats, and both wires sit at 0 volts. When a cell enables it's transmitter, one wire rises to about 4V and the other to about 1V. This is how the bus should look when we send 0xFF:

And here is how it looks when the EVD5 sends 0x55:

Note the difference in the first transition, the the inverting line should go from high to low, but it goes from 0 to low. This is because the transmitter is enabled during the start bit, rather than a sensible time before the start bit.

I tried enabling the bus with a very short pulse (arrowed) before sending a character (note inverting is on the bottom trace instead of the top like above):

This almost worked, but I couldn't avoid framing errors -- the receiver sees the pulse and gets confused. I need to wait long enough for the receiver to assume the start pulse was not the start bit of some random character, but the transmit enable circuit times out and disables the bus before this happens.

There are two possible solutions, capacitive termination, and increasing the baud rate. More on this later.

Recently I visited Sydney and met up with some of the local EV community.

I saw Nathan's workshop at Convert UR Car and he was kind enough to show me a Toyota Yaris conversion in progress and even let me drive Sparky, his lead acid powered '92 Hyundai Excel. I can now say I've handled a germanium transistor, as he had to replace part of the control for his army surplus generator.

I also met up with Anthony Wood (who helped assemble my Mini in January) and Michael Day (who's converting an MR2 with an MES-DEA motor) for lunch and had a ride in his Anthony's (yet to be converted) Cortina.

Jaron Ware & Mark Taylor showed me their Formula V race car which was converted to electric some time ago. When I turned up they were installing an open source DC motor controller, a variation of Paul & Sabrina's design.

Mark also had a very tidy flooded electric van conversion:

Jaron told me stories about his trip to Lake Gardener to crew for the Catavolt team. Kearon de Clouet did 177kph on a Modified Partial Streamlined Electric Motorcycle for a new world record. Besides getting the record, they learnt a lot on this trip, it turns out you should transport your gear 2000km in a covered trailer, especially when the last half day is on a dirt road and it's raining. The dirt was especially prepared by the Australian climate to get into everything. Also, you should get a big trailer and make a mobile workshop, so you can charge and perform maintenance while waiting in the queue for the next run -- you spend most of the day queuing and a mobile workshop would let them get more runs.

Paul Tuffery from the Video Guys came to Motukea from Christchurch and Bill let him drill a hole in the KillaCycle!. These are the best videos I've ever seen of the KillaCycle. In some ways they are even better than being there in person, as he has each run from 3 different angles in glorious HD video. See the KillaCycle channel on Vimeo for 5 more videos!. Make sure the HD button is off if your computer can't handle the large files.

Killacycle. Standing 1/4 mile. 4th run. Motueka NZ. 6/2/10. from Paul Tuffery. The Video Guys on Vimeo.

Athol Williams made a house to keep the KillaCycle dry while travelling on the back of his little Stud Welding NZ truck. We really can't thank Athol enough, he's the New Zealand & Australian record holder on his 6.1s nitromethane powered bike. He took care of an aweful lot of logistics on the KillaCycle tour, and it wouldn't have been the same without him. During the first part of the tour, the KillaCycle travelled in his mobile race workshop, and he provided racing advice and support at Meremere and Nelson. It was a great weight off my shoulders to have his years of drag racing experience, as I could simply be pit crew while Athol took care of the track side with Bill.

We had to wait after a V8 powered mini took out some timing gear in an unscheduled bit of off-road adventure. He managed to get it back on the track, and was loose all day.

Scotty

entertained

with

a

burnout

Not everything went smoothly, with the KillaCycle refusing to switch the motors from series to parallel. The bike starts in series, putting up to 2000A through both motors, but in series, the voltage across the motors adds, so when they are turning fast, we run out of battery voltage to push the current through them. Scotty pushes a button to switch the motors to parallel, where the current divides and we get up to 1000A through each motor, but the voltage doesn't add, so the battery is able to do this even when the motors are turning fast. Unfortunately, the shift only worked on the first easy run and while testing in the pits, it wouldn't work on the track at full power. Without shifting, we only managed a best time of 8.81s in quarter mile.

We did do a quick turnaround for the last run of the day, doing two runs within 8 minutes. We didn't do anything to the bike, we just waited behind two other cars before we could go again. The KillaCycle can do about 14 passes, including riding back to the start line, without charging. Bill tells me that when they do showboat back-to-back runs, the second run is often quicker than the first, but there is a good chance of damaging a motor as they get too hot. This time there was no damage.

Nelson is going well, we've done 2 runs with the KillaCycle, and are about to do a third. The first was a slow 10.551s @ 115MPH where we found the track is very slippery out of the hole. In the second run we had a ground fault in the motors and the series-parallel shift didn't work, so 9.594 @ 136.37MPH was slower than expected.

We hope to do an 8s pass next.

I'm in Nelson where the KillaCycle is scheduled to run tomorrow (Saturday 6 Feb) at 11:30 at Motueka aerodrome. The experimental "high voltage" motors that blew up at Meremere have been replaced with the record setting lower voltage motors with variable brush timing system, so we should see better times than last weekend.

The 4 & Rotary Show and Shine went very well. Come down to Meremere to see the KillaCyle run, we're expecting it's first run to be about lunch time.

Phil & I presented at LCA2010 this morning. The talk went well, but suffered from not doing a practice run. I guess I shouldn't have been writing the slides half an hour before the talk started. I was late getting the slides together because I got the Mini running yesterday evening! I "just" have to install the instrumentation and bleed the brakes and I'm ready for the open day. Video of the talk and the slides should be available soon.

Bill Dubé also arrived yesterday and chatting with him is great. The KillaCycle arrives in Wellington tomorrow and goes on display at Te Papa on Thursday morning. Bill's free public talk is at 6:30pm on Thursday at Te Papa. See Phil's blog for the rest of the KillaCycle tour schedule.

I've now re-installed 30 of the 36 cells in the rear battery box. I added plates at each end of the box to make the walls flat, so it will provide better compression to the cells. This is working better than with no plates (there is a lip at the bottom where the floor of the box overlaps and is spot welded to the sides, but this lip is only 10mm high), but sadly it turns out Auckland is shut in January, and the cutting department is still on holiday. I improvised with 1mm steel (old computer cases turn out to be good for something) and a plasma cutter. Many thanks to Edward for lending me his plasma cutter, it's fantastic, no home is complete without one. I need to reinforce the outside of the battery box, as it bulges a little when you swash the cells in. I realised this only after I'd put three coats of paint on it :-(

Anthony (woody on the AEVA forum) is in Auckland this week and came to have a look at what I'm up to. He was a great help, passing clamps while cutting, sanding the terminals on all the cells and helping me install the cells. He also pointed out many obvious things that I missed (things like you should hammer cells on the edges and the conical plasma cutter torch is conical, but it still has a flat face which you can use to run down the side of a guide and get a straight line).

Sadly the second battery rack was to complex to finish in time for the LCA Open Day and KillaCycle tour so I'm going to have the same 36 cells and 118v or so open circuit that the car had last time it was working.

I've mostly finished the bracket which will hold the front mount (where front is closest to the front of the car if the motor was in a north south configuration, which means it's on the left on this car). The mount is the black round thing hanging below the pipe I welded in earlier, and I've spent the evening grinding the curves on the bracket between it and the pipe. The bracket needs a bit of tidying up I'll be ready to weld it onto the frame.

In unrelated news, the KillaCycle arrived in New Zealand safely, the Tumanako VC motor controller had it's first drive yesterday in the Greenstage electric Saker (first time that has moved under electric power too!) and I have about a week left to put this car together before I'm leaving for LCA2010 in Wellington.

I've mostly finished the rear battery rack. I need to fold up some feet to lift the rack up off the boot floor, decide how to support the front and rear sections (cantilever from the middle one? more feet?) and devise a way to hold the cells down.

For a while, the rack came with added vice grip.

I once said I should get a load leveller. Well, I finally gave in and got two. I think I did well with $50 Chinese units from Save Barn. After a little fabrication, they hang at right angles and I can adjust everything in comfort. I'm never installing another engine without one.

Part of the fabrication involved a strap with two holes, and the levellers came with a piece that just needed a little straightening. I'm not sure if this steel is particularly hard, or just poor, but it did not take kindly to being straightened.

I'm making progress fashioning the structure between the gearbox and the mount that will isolate it from the frame of the car. I'm using an aluminium and rubber mount from a small Nissan (the round silver and black thing between the tube on the left and the gearbox in the middle).

I've been messing around with the BMS master hardware instead of welding stuff together. I'm trying to send a CAN message and, rather than write my own library from the datasheet, I tried to use Microchip application note AN878. It was easy to adjust so it would compile with SDCC but I seem to have found a bug in the compiler.

Because the PIC is a Harvard Architecture CPU, pointers to code and pointers to data are handled differently, which SDCC implements by adding some extra bits to the pointer. Unfortunately it's possible to loose those bits while casting the address of an SFR to a pointer and assigning that to a pointer array element by index.

With the the bits cleared, the write-to-pointer-target helper will do nothing, leaving the pointer on the stack. When it returns, the rest of the code freaks out because it expects the pointer to have been popped off the stack.

The solution (apart from fixing the compiler bug) is to leave out the cast. It seems that assigning an SFR to a pointer array element doesn't actually need a cast.

BAD	pb[1] = (unsigned char *) &TXB0CON;
GOOD	pb[1] = &TXB0CON;

After many hours of fettling and weld practice, I've managed to weld the reinforcing bars into my front subframe. The motor will hang between these bars and they also provide strength at the front of the frame for cells or other equipment.

My 175A welder took a bit of tweeking to make good welds in 3mm steel. The biggest change that made the most difference was switching from 0.6mm to 0.8mm wire. I think this could also have been achieved by turning up the wire feed on the 0.6mm wire. MIG is a little counter-intuitive, if you weld isn't penetrating and the wire is building up in a large bead on top of the work, you'd think that you have to turn down the wire speed and weld slower so the heat has more time to penetrate. Actually you want to do the opposite, as the faster you push the wire into the work, the smaller the arc, the greater the current and the deeper the penetration. Of course wire speed isn't the only parameter which affects the current.