Red Suzi

Post by **acmotor** » Wed, 24 Sep 2008, 06:22

Just some pics....

Red Suzi (1982 Suzuki Sierra)
11kW industrial AC motor with 30kW Danfoss controller and 12kWh of sealed lead acid. 40km range 82kmph top speed. see www.evalbum/1149

Front junction boxes. Water and dust resistant probably IP55 or so all up. No exposed terminals or even fuses. International electrical hazard labelling. All the control and 12V aux charging comes via these boxes and centre of traction battery pack connects to chassis and motor body via 75A fuses in clear box. Sorry, only +-12V inside any of these boxes.
The higher voltages only come together at the Danfoss in the back.

Underbody conduits at rear of vehicle 6 of... -96V, front battery box charger lines, +96V, protective ground, comms to front junction boxes and 12V from DC-DC to aux battery, 11kW motor feed. The build to +-288V is at the rear battery boxes. Wiring is 6 square mm copper.

Original motor fan replaced with 2 x 120mm 12V fans.The original motor fans are designed for bi directional rotation at the sync speed so poor performers and no fan at low revs (where it is most needed) and closer to a siren than fan at high revs.
The original fan cowling and body cooling fins are well designed. The air flows neatly along the whole body travelling in the slots of the cooling fins. Well done somebody at ABB !

Post by **Thalass** » Fri, 26 Sep 2008, 18:18

Awesome. Good wiring setup! I can't stand wires that are unrestrained. eugh.

Post by **TropicalEV** » Sat, 27 Sep 2008, 02:50

Nicely laid out...Well done...looks very professional and a serious conversion. are the motor mounts on rubber?

Post by **acmotor** » Sat, 27 Sep 2008, 05:08

Thanks for the positive vibe.

Motor 'cradle'(light grey in pic) offers up to the original engine side rubber mounts and gearbox rear rubber mount. (actually new rubbers from suzuki dealer).
Cradle is welded steel angle.
The suzi, being 4WD, had quite substantial engine mounts that have hard bottom outs in all directions other than 45deg to horizontal from each side and this direction of movement is not possible when you think about it. The emotor weight (80kg) is slightly less than the original ICE and G/Box so it works out well. It can only move about 15mm in any direction.

The body of the motor is quite deliberately connected around the rubber mounts by 6sqmm earthing cable to the vehicle chassis for electrical safety. (same as the body of all electric motors are connected to earth in mains wiring).

Post by **acmotor** » Mon, 29 Sep 2008, 05:02

How about a reverse switch ?

The red, high intensity led below switch lights up the cab at night !
The switch is 90deg rotary so operation needs to be quite intentional.
If you select reverse when going at speed forward, the controller will take 3 seconds to ramp up any reverse torque (set by controller settings) so nothing violent. After that the regen braking will pull you to a holt, then you get reverse (all that if you have accelerator applied that is, otherwise nothing happens).
PS. There is no delay if you are at zero speed in the first place of course !

The switch is DPDT and also directly operates the reversing lights and annoying beepy thing !

Post by **moemoke** » Mon, 29 Sep 2008, 18:41

Great work there Acmotor,
do you have a blog or build info with your Suzuki, looks like a well thought out system, would like to do the same in something a bit more aerodynamic.
What speed and range do you hope to get?

Post by **woody** » Mon, 29 Sep 2008, 21:59

moemoke wrote: Great work there Acmotor,
do you have a blog or build info with your Suzuki, looks like a well thought out system, would like to do the same in something a bit more aerodynamic.
What speed and range do you hope to get?

48km

A moke is very aerodynamic with a toneau and no windscreen

16kw should be enough for a moke

cheers,
Woody

Post by **acmotor** » Tue, 30 Sep 2008, 06:58

No time for blog !
Early listing is on evalbum/1149
I'll collate more info and put it up sometime.

Just drove 37km trying out PC connected to controller. Free software only logs 1000 points (60 seconds). I'll have to buy the full version.
Used 5.2kWh off batteries (probably 7.2kWh is all I can take of SLAs for 100% DOD) so 37km is around 70% DOD.
Removed the front tail shaft so I could try low and high ratio on bitumen.
Leaps to 30kmph in 1 second in low ratio but top speed a bit over 40 kmph. I'll stick to high ratio before I twist an axle.

The only thing aerodynamic about the suzi was the donor price ! The rest has the ballistic co-efficient of a 44 gallon drum !

Post by **woody** » Tue, 30 Sep 2008, 16:03

acmotor wrote: Just drove 37km trying out PC connected to controller. Free software only logs 1000 points (60 seconds).

Cool. Can you send me the log of a flat out standing start ?

Post by **acmotor** » Tue, 30 Sep 2008, 17:37

?? Data file contains only 1 channel - motor current. Free software !!
Not much help - early days yet ! At least the data is starting to flow.
I'm onto Danfoss for the full software.

It seems to have recorded 20 seconds at around 7 samples per second.
Just enough to wet the appetite !
Motor current is absolute only - no indication of regen in this data.
Run was accelerate moderate to 40km then stop, accelerate hard to 40kmph and hold. ( in a deserted car park ) I didn't get to log any 60kmph runs as I was still getting the software to run. It was a demo version and wanted to demo not run !

suzi_run_1_motor_current.xls

Post by **acmotor** » Mon, 06 Oct 2008, 23:01

Motor cradle view from under vehicle looking forward from transfer case.
Original rubber mounts are used.

Number plate 'EV' label. (Better than an LPG one !)

Post by **antiscab** » Tue, 07 Oct 2008, 04:20

rubber mounts? oh no....u havent crossed over to the double isulation side have you?

Matt

Post by **acmotor** » Tue, 07 Oct 2008, 05:04

Ha Ha, trust you to come up with that Matt !

No, I do not pretend that rubber mounts are insulators, with a big metal motor there for all to touch !
Try finding the data sheet from the vehicle manufacturer that states that the rubber mount is designed for 300VDC (good luck). I know some people 'imagine' double insulation. I don't. My emotor body is securely connected electrically to the chassis as it should be. My electrical wiring and terminals are where the double insulation (where practical) is applied. The motor itself is TEFC to IP55. This satisfies NCOP14. How about your motor ?

I used the original engine and gearbox mounts so there is no question from DPI about their suitability to carry the emotor.
One less engineering report.

It would almost be possible to do away with vibration isolating rubber mounts with a (well balanced) emotor, but VSB14 section LA that my DPI approval letter requests that I comply with states ...
2.3 "The engine mounts must provide for vibration isolation between engine and body"
This may be a bit historical and ICE age but there it is.

Post by **acmotor** » Thu, 09 Oct 2008, 04:03

The engine bay is starting to look quite tidy.

These two tool boxes from Captn. Bunning have 12mm ply floors and laminex linings added. They are bolted down to 25mm x 3mm angle iron frames that are bolted to the vehicle. Hold down bars across the top of the batteries go down through the bottom of the box and into the angle frame. They are about 39kg each all up and hold the +12 to +96V and -12 to -96V sections of the battery pack. Each box contains 1/6 of the total vehicle battery pack. The hold-down bars/bolts and angle frame base are the real structural items. The metal box is to keep it electrically and fire safe !

Battery pack 3 front, side (+12 to +96V). The battery pack breakup contactors can be seen mounted to the hold-down bar. The box and this bar are securely earthed ( connected the vehicle chassis ) so they cannot become live due to a fault - just good electrical practice.
When the contactors open by Estop/ ign. switch or inertia switch, the pack is broken into 48V floating modules. Pack can thus be charged as one 48V unit via multipin charger plug. Boxes can be locked for safety and there are no exposed wiring or terminals outside the box. Conduits are orange, wires are not. The PCBs are the simple zener battery equalisers.

Post by **Richo** » Thu, 09 Oct 2008, 05:29

Very nice job

Any pics of the controller placement yet?

Post by **acmotor** » Thu, 09 Oct 2008, 06:46

EStop and Danfoss LCP (local control panel) mounted on dashboard.
Estops are double wired to Danfoss Estop and trip lock signal (N.O. contact) and 12 x battery breakup contactors (N.C. contact) (this is fed via ign. switch so that cuts everything as well - as it should)

RHS of Danfoss mounted over rear battery boxes. Wiring covers removed. Small DC supply for battery current transducer showing and rear Estop.
There are 3 Estops in the vehicle. These also double as 40g inertia switches (front direction only). Front switch is in crumple zone so 40g is possible. If others see 40G then you won't remember it !

Rear battery boxes open and being 'megga' insulation tested (box with knobs in centre of pic.)(500VDC to chassis and across contactors when in breakup mode).
Danfoss is sitting up on 450mm 'service' brakets to allow battery box to open. It normally sits down just above box lid.

Post by **BjBlaster** » Fri, 10 Oct 2008, 05:35

acmotor 12 x battery breakup contactors (N.C. contact) (this is fed via ign. switch so that cuts everything as well - as it should) wrote:

Just curious about this bit... does this mean you are energizing the 12V contactors to isolate them as they are NC normally? What if your 12V coil supply is damaged? You wouldn't be able to isolate then would you?

Post by **acmotor** » Fri, 10 Oct 2008, 06:44

Sorry Bj, I probably tried to fit too much in one sentence.

What I said was...
Estops are double wired to Danfoss Estop and trip lock signal (N.O. contact) and 12 x battery breakup contactors (N.C. contact) (this is fed via ign. switch so that cuts everything as well - as it should)

Perhaps if I clarify further... The Estops have two separate internal switch units (telemechanic brand) One is N.O. and one is N.C.
The N.C. contact of the Estop is considered the failsafe closed circuit for energising the battery breakup contactors. When the Estop is NOT operated (pressed) the contactors can close with 12V via the ign switch. (The battery pack contactors are N.O. of course !)
The N.O. contacts of the Estop go to the Danfoss "Estop and trip lock" signal inputs.
Two stop signals (from 3 Estops) are better than one !

This is front (engine bay) Estop showing double wiring of two stop functions.

I note commercial hybrids (wash my mouth out) only use contactors operated via ign key (and inertia / fault systems) electrical No Estop and definitely no mechanical (direct acting) Estop.
There is usually a service switch at battery pack as well that requires the removal of back seat or boot lining and a screwdriver to access. Not meant for user to operate and not suggested in guidelines for emergency personel to operate.

I note also that Emergency Response Guides for hybrids and Fire fighter / vehicle extraction service instructions are to double CUT the leads to 12V system battery of a crashed vehicle after first removing the ign key. Negative first then positive so that all electrical systems will then typically shut down. This would not be the case with a mechanical Estop ! Get with the times NCOP14

For a converted EV I would use plenty of enthusiast operatable electrical Estops around the vehicle. Good safety practice.

BTW if anyone doesn't know why double cut, then ask !

Post by **BjBlaster** » Fri, 10 Oct 2008, 18:46

That is neat. I like the idea of using a low voltage estop, but how will this go passing NCOP14? I want to use 2 contactors to break my 96V up into 2 48V strings (front and back) - but this isn't preferred as it must have direct mechanical isolation which doing the above doesn't comply with. Is this going to be revised any time soon? Especially when it says to keep the high current cabling out of the cabin, but you will need it there if you have to have a direct mechanical switch (or use a choke cable and a breaker from under the bonnet).

Either way your red rocket is looking great! Excellent cabling detail and I hope to see it on the weekend

Post by **acmotor** » Fri, 10 Oct 2008, 20:23

NCOP14 will move on IMHO. It serves us well.

There is commercial precedent for electrically operated battery pack contactors by Toyota, Honda and Ford that I have looked into. No 'Estop' buttons either !
These involve at least two contactors and integrate inertia systems and earth fault systems.
The ign key is the first port of call in any case.

However, on the EV conversions scene....
Estops are smart safety items.
I consider they should remain (in electical form, perhaps with double function).
One should be located within reach in boot, engine bay and cab.
A mechanical switch, removeable plug or fuse could also be in the system but comes a poor second after a battery pack broken up by multiple contactors.

BTW,
It seems the original idea of the mechanical stop switch in NCOP14 was to (only possibly) get around the rare case of welded contacts in a contactor. If a contactor is used at design rating this is rather rare but still possible.
If there are two (or more) separate contactors then not only is the voltage across each set of contacts less, the chance of welded contacts on both units...... at some point we move on !

Post by **acmotor** » Wed, 29 Oct 2008, 07:43

Charging connection box. In the ute back between rear battery boxes.(Danfoss raised up for this photo)
Two waterproof 16 pin connectors connect charging options of 1 x 48V up to 12 x 48V SMPS chargers or solar panels to traction battery.
Connectors have interlocks to battery breakup contactor system.
Not bad being able to charge the 576V suzi off 48V solar !
When ign is off or ES pressed the max voltage anywhere in the vehicle is 48V (nominal).
This includes during charging. Safe move.

Inside the charger connection box where all the 48V module charging lines come together. Each line is fused by 3A fuse at each 48V battery module and also feed through 3A anti-backfeed diodes at the module end.
So there is not much potential for trouble in here.
The left side (upper in picture) is for the +ve batteries and the right side for the negative batteries. There is a divider in between. Just to keep max separation between the voltages when contactors are in.

Post by **acmotor** » Sun, 09 Nov 2008, 05:44

Woody or Richo,

run6_5km.xls

If you get a chance, here is a data file from a data logging run in red suzi today with the full version of the Danfoss software. About 5km long. Starts with backing out the driveway.
Only 10% regen enabled, speed control open loop mode.
I played with all sorts of graphs until I was left with plotting my coffee temperature ! Some things are quite predictable. Others leave me wondering. Anyway, have a play with the data. It doesn't get any more real.
Any complaints - A was driving - OK.

Post by **a4x4kiwi** » Sun, 09 Nov 2008, 15:04

Hi Tuarn,

Where is your 'current sensor' data coming from? I had though the VFD only output absolute values.

cheers,
Mal.

Post by **acmotor** » Sun, 09 Nov 2008, 18:35

Somebody is paying attention !

Mal, as you know, DC bus current is probably the only measurement that, although Danfoss probably make it internally, doesn't actually appear in the long list of user readouts.
This DC bus current (amps from battery in our case) is a need to know number in EV land so I added a LEM 200A current transducer (clamp on on the battery wire) and read it into analogue input 54 of the controller. This input is of course loggable. Input range is +-10V . LEM range is +-5V for +-200A ... that's fine. Scaling factor is 40 (5V=200A). This I applied in the excel file to create the battery current column.
Resolution is 10bit + sign ? So this gives +-512 bits (+-5V) for +-200A = 390mA
I did use -40 so the signs matched when looking at the data since I had placed the clamp on so battery drain was -ve. I thought that was logical at the time !

The 2 samples per second update rate is probably a bit slow. Actual rate is number of channels dependent. This was the max for 8 logged channels. You can't read too much into the short term trends.
The small (10% regen used on this run) is evident as is the near 100% efficiency of the controller.

On the 5000 series, the only absolute value data is the kW to motor.
This annoyed me at first but by looking at torque direction (+-) and battery current (+-) it is now possible to separate out the regen component.

Integrating the speed column wrt time to give distance shows 4.8km so this looks correct.
Integrating battery kVA (kW) shows about 0.97kWh nett.
Looks about right, given this was very stop/start and playing around.

I'll probably start again with controller settings now that I can see what is going on.
BTW, if your blood was only the temperature of the motor after this run, you would have serious hypothermia ! The computer fans work fine.

Post by **acmotor** » Tue, 11 Nov 2008, 06:00

A section of logged data, about 3 minutes worth, showing a take-off uphill with plenty of torque on the motor. Remember this is direct drive.
Regen limit was set at 20% at this time and this can be seen in the negative motor torque.
I'll wind the regen back up on the next run.