Vogue on Display Yesterday

The Vogue was on display courtesy of the Sunbeam and Talbot Owners Club yesterday at Federation Square in Melbourne. I had a lot of interest in the car, such that I hardly got time to eat lunch (Hi to those folk who scribbled down the blog address). This was the first time that the Rootes Group and Humber Car Clubs of Victoria had seen (or heard of) the Electric Vogue as well.

We did an FAQ and a basic diagram to stick in the windows but there were still a ton of questions.
Some people were amazed that production Battery Electric vehicles were available to buy. Others were "stoked" that this could be done to an old car. It was a very positive experience for me.

Thanks to Matt Lambert for the picture - this was a quiet time. Often there was a crowd where I had to shuffle in and show them the spare cell and explain some of the stuff the FAQ didn't cover. (I'm the one with the terrible posture.) One bikie looking guy had a quick look at the engine bay then looked quickly around and sighted me. "You must be the guy that built this" he says. "Why did you pick me" I asked with a smile. "You just look like the guy" he said. Hmmm.

 We all lined up at the Russel Street extension until they let us into the square. In the picture below, the Vogue is two behind the bright blue Super Minx Station Wagon (a bit hard to see). Hillman and Humbers on the inside and Sunbeams on the outside (mostly).

From the other side.

The highlight of my day was when the Fed Square people came around with a stack of oil drip trays and I got to say "I don't need one of those".

Some Vibration Testing and Thoughts

I have added a new search label to the blog - Vibration!

Someone on the AEVA forums had suggested that the vibration may be resonance in the motor mounting system. I had my heart set on this being the cause - alas no. I'd even measured up for motor cycle steering dampers.
Placing a jack with a block of wood under the centre of the motor frame very marginally decreased the vibration is accord with what you would expect in that it offered a bit of dampening to the real cause. I used my phone to record the subsonic audio coming from the car during axle stand testing. Here is a spectrum plot of the recorded audio with speed somewhere between 55 and 60 km/h.

The lowest peak occurs at about 76 38 Hz which is rotation frequency at 57 km/h - that fits. The next peak is at twice rotation frequency which suggests the universal joint as it rotates over the "top" on each side. The third and most prominent peak is at 118 Hz - which I can't explain - three times rotation frequency?
It still looks like interplay between the UJs and the motor/mounting.

Crawling around under the car travelling at 55-60 km/h (on stands) the vibration was almost non-existant at the diff end. The motor itself was the worst point. I could not visually see any runout consistant with this behaviour. I've had some really bad UJs in tailshafts in my life that didn't vibrate as badly as this apparently OK one.

Precision Balancing have a portable setup for analysing vibration that they use in the field - it may soon come to that.

I'm also thinking about CVs....

Vogue on Display this Sunday

The Vogue will be taking part in a Rootes vehicles display this Sunday at Federation Square in Melbourne. The display starts at 10AM and goes until 3PM. If you see me there, please say hello.
I apologise in advance if something unexpected happens and I don't get there - it's unlikley since I drove the Vogue for two weeks without any car-stopping incidents.

I'll try to post here when I'm there on Sunday.

Economy for Last Week and Full Current at Stall

Power economy was pretty similar to last week.

Week ending 19th April 2013:
Monday 11.86 for 48.92 km => 145 Wh/km
Tuesday 10.1 AH for 48.1 km => 126 Wh/km
Wednesday 7.33 for 34.12 km => 129 Wh/km
Thursday 9.93 AH for 39.5 km => 151 Wh/km
Friday 10.75 AH for 47.83 km => 135 Wh/km

All battery to wheel.
Thursday was the day I recorded serial logs. Typical that it was the worst day.
The Vogue has now traveled 636 km as an EV.

Friday I played around with different PWM frequencies. I hadn't realised that I wasn't getting full current at stall until I tried 4 kHz flat-top (see previous post). Suddenly I got good acceleration from stationary. The only problem is that 4 kHz is annoying. I tried a few combinations during the day and was only on the last trip to Doncaster (up a significant hill) to pick up William (son) that I tried the last pre-supplied switching frequency combination.

It's 8 kHz SIN auto-switching to 2 kHz  when current exceeds the 8 kHz limited value.
It works perfectly. I had full current from standstill, same as the 4 kHz configuration but no noise when not heavily accelerating. Unlike the 16/8/2 kHz configuration, it apparently applies the correct current limits when switching PWM frequencies.

It made a huge difference. There was a small hill turning out of a driveway that William and I had always wondered if the Vogue would handle once it was going (challenges of direct drive). It was no problem. It may have been another story with the 16/8/2 kHz configuration that I have been running since the car was on the road.

I'm not driving the Vogue this week. I have decided that I'm shaking it to bits (a couple of the chargers have rattled their voltage output down by half a volt or so - trimpot movement) and I need it going for Sunday. With Thursday a public holiday here, I'll do the axle stand thing again and check out Richo's (AEVA) theory that it may be resonance in the motor cradle. I'll also be a lot more critical of which end the vibration is coming from.

Second Week as Daily driver

Not much to report about the Vogue other than I have driven it every day this week with no issues. I logged data from the dashboard to and from work on Thursday and some interesting things came out of it. The major one being that my acceleration up to 4 km/h is slower than 4 km/h to 70 km/h. This has to do with the controller de-rating it's output below 5 Hz.

This isn't a hard problem to get around - in fact by changing it's "carrier" or PWM switching frequency to 4 kHz the problem goes away - but it's too noisy. I currently use 16/8/2 kHz auto switching where the controller auto-changes based on motor current but it appears to have an idiosynchrosy that retains the 16 kHz current limit even when it's auto switched to 2 kHz. It looks like I'll have to add some function blocks to switch it over.

My logging laptop on the passenger floor.

A timer on the carport 15 A power point would be handy too. I currently go out each night at 11PM to turn my chargers on. I have bought a 16 A timer in a nice outside enclosure but I have to add some leads and install it. Soon....

I'll post the daily economy figures for this week during the weekend.

Heater Failed this morning

Ah prototypes...
Since it's the start of the cold weather in Melbourne, but not TOO cold yet, I thought I'd better run the heater for more than a few seconds. So, this morning about 1 km from home I turned on the heater, directed the airflow to "screen" and put the control about halfway on. The screen cleared almost immediately (it wasn't bad to start with) and the cabin warmed noticably within a minute or so.
I figured I'd save range so shut off the blower (fan). History: The heater PWM controller runs from the +12VDC feeding the fan, so when the fan is turned off, the heater controller stops feeding pulses to the IGBT and the heater core stops getting any power - or that's what SHOULD happen.

What actually happened is that a slight smell became evident in the cabin so I quickly turned the blower back on. I was stopping for a small errand in about one kilometer so I didn't pull over just then. Once my errand was done (just a pick up) I started the Vogue (closed the 600 V contactors) and saw that there was a 1.6 A current draw. I shut it off again, opened the bonnet and unplugged the conveniently located heater 600 V  power wires (not sarcastic - I didn't trust the heater controller as you can see from other posts about "heater"). Such was my distrust that I hadn't even mechanically anchored the wires yet so it took all of 5 seconds to unplug them.

I'm glad they are finger proof (heatshrunk Anderson PP35 connectors) because I totally forgot to wait the 30 seconds for the controller bus caps to discharge so they were essentially "live".

So it appears my PWM controller experiment needs more work.
For the technical folk reading the blog  - my heater controller circuit.

It probably needs:
1. Zener across IGBT gate - I thought they were built in - it seems not.
2. Larger bus capacitance. 1uF is probably extremely optimistic.
3. Better IGBT heatsink. I used a little flag one assuming the blower would keep it cool. (The control board is mounted on the blower/inlet side of the heater core).
4. There is also no negative voltage on the IGBT gate for safe switch off - hmmm.

A lot of these decisions were based on me thinking that the Ceramic heater core would not be very inductive - I need to test that assumption. The 1.6 A current draw suggest that the IGBT is NOT totally shorted - more dangerous than if it was (fire hazard). It will be interesting to see what is really going on.

The heater was finished before I had 600 VDC to test it so this time it will get a full test before re-installation. It's pretty easy to get the heater box out - no dash disassembly or such involved and everything just plugs in (famous last words).
I shouldn't need the heater for a couple of weeks yet so it may have to wait until after the 28th April when the Vogue is involved in a display.

**Addition 16-Apr-2013 I should also slow the PWM switching frequency down. 16 kHz is not necessary. I think I'll slow it down to 100 Hz. This should remove the necessity for some of the other possible changes - bigger caps, negative gate voltage.

Economy figures for the week ending 12th April 2013:
Monday 14.1 for 58.8 km => 144 Wh/km
Tuesday 7.22 AH for 32.06 km => 132 Wh/km
Wednesday - forgot to record it.
Thursday 8.6 AH for 39 km =>  132 Wh/km battery to wheel.
Friday 11 AH for 50 km =>  132 Wh/km battery to wheel.
Saturday 2.3 AH for 9 km =>  153 Wh/km battery to wheel (three 3 km trips).

Charger Mods Done

I have finished modifying all the chargers and was pleased to come out this morning to the Vogue with every pack charged and fully balanced. This is the first time all the packs have been fully balanced at the same time and will be the norm from now on.
I had yet another fuse holder fail Wednesday night sometime and Pack #1 (boot) didn't charge for my drive to work on Thursday. I bypassed it at work and charged just that pack - so it didn't stop me.

I just have to get rid of those pesky fuseholders now.

Robert - you have a fan!

I generally don't place any non-Vogue related info on this blog and also dislike "hero worship" - it gets in the way a rational thinking. For this man, I make an exception.

Just about everything he writes or says is something that I feel strongly about.
His name is Robert Llewellyn and is perhaps best known for the Kryten character in the BBC's Red Dwarf series. He is a supporter of EVs (see the youtube series Fully Charged) and a person that makes the world a better place because he is in it.
Here is a link to his blog.
http://llewblog.squarespace.com/

Third day as daily and Battery Alarm sounds

On was on my way home last night and stopped at a major intersection to turn right. When I got the green arrow and moved off, the battery alarm alert started. I pulled over as soon as practical and popped the bonnet. (I do not yet have my controller keypad in the cabin which tells me which of three regions in the car the alert came from). All the blue lights were still lit on the engine bay packs. I moved around to the boot - and it was LOCKED. So I turned the car off, unlocked the boot (I needed the key from the ignition) and got back in and started driving again. The alert started within 10 seconds. I stopped again and checked the boot packs. Pack #4's light was off.

At this point I should have bypassed pack #4 which would have taken all of two minutes but for some foggy reason I decided to limp the 3 km home.
I got home OK (annoyed a bit of traffic) and checked voltages on all packs. Most packs were 52.4 VDC. Pack #4 was 36 VDC. In my mind I had clearly ruined some cells.

I placed the car on charge and within a few seconds the charger light for pack #4 showed it was fully charged. Hmmm. To cut to the chase, the inline fuse from the charger to the pack was very hot and the fuse holder was melted and at a strange angle.

The diode isolator and inline fuse assemblies for the chargers.

This is the second fuse holder that has melted - I replaced the one on pack #12 about three weeks ago. These fuses only take 3 Amps but it appears to be too much for the el-cheapo fuse holder.

I pulled the diode/fuse assembly out and soldered in a fuse directly and heatshrunk it with a couple of layers. I then placed pack #4 back on charge. It didn't get hot at all after that. Pack #4 took a full 3 Amps for more than 6 hours so it clearly survived and I learnt some helpful clues.

Clue #1
According to the smart meter online monitor, the night after my 60 km day, the house drew 3 kW for an hour then 2.7 kW for 3 hours. The clue was that one charger had dropped out after the first hour (smartmeter resolution is 0.1 kW).

Clue #2
On the drive home, several kilometers before the alarm, the full pack voltage was dropping under 600 VDC regularly. I had not even seen that on the 60 km run, except under really heavy load. Now I know to pay attention.

Clue #3
No so much a clue as observation. Don't skimp on parts - especially cheap ones. I'll replace all the fuse holders as time permits and solder the fuses directly into the assemblies.

There are obviously down sides to the one-charger-per-pack system but the alert system obviously works fine. I need a reliable wireless mains power monitor for the EV outlet - the one I have is rubbish due to the non-linear power factor issue with the chargers (I think I'll crack the sender open and see what I can change/fudge).

The battery alarm drives you insane - it's very loud. I was originally concerned that you wouldn't hear it because it's up under the dash somewhere. The alarm also locks in and can't be cancelled without turning the car off - not hard to change but I'll see how it goes - It normally shouldn't go off anyway..

I am now carrying a short bypass cable and the tools to bypass a pack (and a multimeter) - the Vogue can run on as little as nine packs (I have twelve).
Yesterday's battery to wheel was 7.22 AH for 32.06 km => 132 wh/km
I drove the Vogue in again today.

Second day as Daily Driver

I drove the Vogue to work yesterday. By the time I had visited the driveshaft balancing folk and done a joyride or two, driven back home, then taken William to his activities and picked him up again, the Vogue had travelled 58.8 km on 14.1 AH. That equates to 144 Wh/km (assuming 600 VDC).

Today I braved the tollway and went on Eastlink for about 4.2km. Other than the driveline vibration, it was quiet and easily reached 100 km/h. I was surprised that it only needed about 12 kW to maintain 90 km/hr. Entering Eastlink, I accelerated up the on ramp and projected range was 58 km by the time I reached 90 km/hr. Two sections later I de-accelerated down the off ramp (Ferntree Gully road) and the projected range was now 66 km. The range estimate was clearly influenced by the heavy acceleration and upwards ramp getting onto the tollway.

The combined Trip and projected Range is currently sitting at about 75 km after the trip to work. Curiously it exceeds the spreadsheet projection which indicate I should have used 3.77 AH - I have only used 3.2 AH. I didn't travel at 100 Km/h for more than a few seconds - 90 km/hr was the traffic speed in the left lane.

A current picture of the Vogue (well almost - it's got a tiny little driver side mirror now). William took this picture for a Coffee mug he had made for my Christmas present. It was a heat sensitive surface where this picture and another of the pre-converted car emerged when it got hot.

The current state of the Engine Bay. I have left this picture at full resolution.

All Together Again

The Vogue is all back together. I wish I could say no more vibration but it's still not good. Anyway the Vogue will be my daily driver from now on so I hope to sort things out over the next few weeks while I'm driving it.