Kelly KDH14651B post-mortem

There were quite a few things happening since last time and I was so busy working on them that I could not spare time to update the blog timely. My excuse finishes here :)

OK, the main thing was that my Kelly controller KDH14651B blew up in smoke. The good thing about it that after all reading all comments from around the net I was kind of expecting it. In a way I felt a relief that I don't have a doubt about its reliability.
Actually it happened at the end of September when I was drive-testing my HR-EV. I was driving it around for a day to test the power consumption and battery capacity. After fresh charge it scored 53km and 203Wh/km consumption in Vilnius city-like driving profile (quite a few ups and downs) which was quite good comparing to standard 250Wh/km as a benchmark. The main parameters during this event were following: speed ~70km/h, slight up-slope driving for around 2-4 previous minutes, battery voltage ~124V, battery current ~180A, controller heatsink temperature ~+29C. So it blew up while continuously delivering around 22kW power for last few minutes. Just to note that the highest power possible I recorded with Kelly was 43kW.
Kelly were not willing to provide me some money compensation. Only offer was a notable discount on new controller which would be basically the same one as just blew up. This was not an acceptable option. Here I'll put some photos what I found investigating its failure.

Disclaimer: Of course it is not proper engineering failure analysis so I reserve my right to be wrong at some of my conclusions. I am basing my findings on my own electronics knowledge and experience.

First I took out my control box from the car. Looking closely you may see the Kelly's cover is slightly popped open from inside explosion.

And here is the photo to see it closer. The cover was just glued to aluminum chassis.

When taken the controller out of my control box I could open the cover more widely. A lot of black ash inside.

Some ash left on inside of control box where the controller was sitting:

Of course the controller like many products of China was built in such way that it would not be possible to disassemble normally for repair. That is a reason that I have not heard of any repaired Kelly Controller. You just scrap the old one and pay Kelly for the new one. That did not stop me :). I sawed off the two sides of aluminum chassis to split top and bottom parts apart to have access to the whole board. And below is the photo how it looks with the top off.

Hmmm, first impression is bad. Not just because of black ash all over. The layout of components is really bad if you are dealing with multi-hundred A currents switching. The electrolytic capacitors sit on one side and power connectors on the other with many switching transistors in between. This layout is a recipe for uneven current and voltage spikes, temperature distribution and stress on the components. There are 42 IRFB4227 transistors in this controller connected in a half-bridge configuration: 21 on lower arm and 21 on upper. The basic rule of paralleling is to keep the conditions as equal as possible for every paralleled part. This rule was clearly taken carelessly.
In addition to that there is really "interesting" solution to increase the power of the controller taken closely in the picture below.

As you can see there are two 10x1.5mm copper bars connecting the bigger and longer row of power transistors on the left to smaller row on the right. This gives me a clear hint that originally the controller was designed to handle maximum currents of up to "400A" and needing a more powerful model they added this section to extended the current by ~200A-ish to total "650A". That would be OK if design approach was taken properly and wide connector plates were used to ensure small inductance and, again, even distribution. Having a long bar with 15mm2 cross-section would certainly not help to achieve that. In hundreds Amp currents every 0.1uH counts.

Closer look at possible failure cause indicated that the epicenter of explosion was first transistor which was closest to bar connectors. Based on previous observations this surely looked like a most probable place to fail: closest to the load and furthest to filter circuits.

It was clear that this transistor was producing majority of bang and smoke.

 The PCB track beneath it was also severely damaged. I reckon that after it's failure all other transistors failed in avalanche fashion which is quite normal for paralleled designs. Out of 42 I found possibly 3 that were showing good initial parameters.

Another point is that it is very important to place good filter capacitors on power lines of such switching devices which could help to reduce voltage spikes and even the conditions for paralleled designs. For that you would need good film capacitors. I didn't find them. I only found ceramic SMD capacitors stacked in several places. These usually not good enough for high current spikes filtering.

The next thing was looking at the attachments of these transistor for heat transfer. For that I was taking off bus bars from the transistors. The bus bars are 3x16mm giving 48mm2 cross-section. This is usually good enough for transferring currents from A to B of 600A from which could potentially be drawn from battery. But on controller itself this should be by a margin bigger to allow more even conditions distribution (again) due to smaller resistance and inductance. Anyway I found that each 7 H-lower arm transistors were attached on L shape 3mm thick aluminum plate which is in turn bolted to the chassis base through PCB. Notice anything wrong? Yes, the PCB was used as a heatsink transfer element! OK, the PCB has perforation with metallisation to build many heat transfer channels but still it is not sufficient to drain the excessive heat from transistors down to the base plate at good rate. Below is the photo of PCB what I found under plate and bus bar.

This makes the excess temperature be the main contributor to failure. Remember, I was driving for few minutes in slight uphill. This allowed the transistor temperature to rise but transfer to baseplate and heatsink was not sufficient to keep transistor in normal operating temperature range.

Here is how the PCB looks from below.

Taking into account MOSFET's increased heat dissipation at increasing temperatures this heat transfer certainly does not look enough to me.

Here is how the aluminum case base plate from below PCB looks.

After these findings I certainly would not buy another one from Kelly even at half the cost because I know it would blow up again. It might work well in golf-kart or similar but it is not suited for high currents.

After that I left without controller. As there are no acceptable options (reliable, not overpriced, flexible, powerful enough, with easy integration, available within 1-2 week lead time, etc) I decided to build my own which would be based on IGBTs. But that is the topic for different post...


Motor RPM Sensor

I have built motor RPM sensor from Melexis MLX90217 chip and door magnet. The Melexis chip was ordered from DigiKey a while ago and cylindrical magnet has taken from door magnet and reed switch pair which is used in house alarm systems.
I made an enclosure of the sensor from 30x30x2 square steel tube of approx 60mm length. I welded on the 2 mm sheet cover on one end and two holder arms with holes on other end. Then made a 14mm hole on one side and welded in a piece of steel tube. Then I glued in the magnet with MLX sensor on its end directed to the inside of the box with three short wires running out to the outside. These wires are connected to shielded microphone cable which is running from the sensor to control box. I've painted the sensor with hammered black Hammerite, let it dry for a day and attached to motor on accessory axis side. Here are couple of photos of finished sensor. Sorry - didn't have time to capture whole manufacturing process.

The sensor serves second purpose too - it covers the shaft from dirt and moisture to prevent its corrosion.


Intermediate Shaft Holder

HR-V's ICE engine had a fastening place where the front left driveshaft's intermediate shaft was attached to with three bolts making sure that intermediate shaft sits firmly in it's place in gearbox and is in steady position at the junction with front left driveshaft.

As the ICE is gone this intermediate shaft did not have its attachments position so it had to be manufactured. I took 2mm steel sheet and welded a sort of box with on side concave following Warp 9's contour and the other side flat with three M10 bolts welded in for intermediate shaft bracket attachment.

On concave side there is one hole. This hole is used to firmly attach the holder to the Warp's 5/8" lift eye-hole with one bolt

 I painted the part with hammered-black Hammerite, let it dry for a day and attached to the motor and intermediate shaft.

Now this intermediate shaft is not going anywhere.


Rack for Control Box and Charger

Behind the scenes  I was doing occasional test drives of my HR-EV for last few weeks. I don't have video of myself but I have video of my wife's EV-grin when she was first driving it which I'll edit and post it at some near time.
Testing was involving a lot of wiring and parameter tweaking to make more and more features of BCMS and car working. For temporary drives I placed control box on two wooden blocks across longerons of the car which worked quite ok but I don't have the photo of them - sorry :) Then I got to the point when I needed to have control box and charger fixed under the hood firmly for longer drives.
To fix control box and charger firmly I designed and built the rack from stainless steel square tubes. I didn't prepare any drawings or sketches for it since it seemed to be quite trivial task from design point of view.
The rack is basically two 30x20x2 stainless steel tubes across car's longerons connected with two bars and fastening loops for bolts at the end to fix it to the longerons. I used 3 stock bolt places on the car's body for fastening the rack and I had to weld in the fourth nut place where no other suitable option was possible. I drilled and welded in 8 bolts for places where Control box and charger will be fixed
The rack came out quite nice. Here are some photos of it.

The control box is almost finished. Next I'll need to make a cover for it. As you can see at the bottom on the sides are two bars with bolt holes for fastening control box to the rack.

The end of the box has all the connectors and cables. I can disconnect all the cables and take the box out for maintenance of itself or the motor underneath.

Here is the picture of rack placed with charger on it.

And here is the rack with control box on it. Of course control box and charger will normally sit together.

The whole construction came out sturdy but at the same time easily dismantable if maintenance would be needed. The intervention into chassis structure is minimum.


Building Electric Heater

Hello folks! It's been a while I didn't post any updates. But that does not mean I wasn't doing anything. There are many things that you need to take care of and building a heater for coming winter is one of them.
The heater cores are taken from cheap household ceramic heater-blower.

I've dismantled two heaters to get two ceramic heater elements as I learned from other EV builders that one element is not enough. Here is the picture of them side by side.

I needed to get the HR-V's stock heater radiator out of the car to measure it's dimensions and to build an adequate substitute. It seemed to be a fairly easy task but I found out that I have to completely dismantle cabin dashboard to access the ventilation box, take it off and get the heating radiator out.
Here is how the car cabin looked when I finally took the ventilation box.

Once I opened the ventilation box I got access to liquid heater radiator.

I placed 2 ceramic heater elements on liquid radiator where the best airflow is expected and marked the dimensions outline of them so I would know their location when building heater elements box.

Next I took 0.8mm stainless steel sheet and built the heater box by marking, cutting, bending and riveting it to resemble the dimensions of stock liquid heater radiator. To hold the elements I used the plastic holders that were originally in electric heater. I sawed them off to take the shape and fit side by side. It fastened them to the box using two bent steel sheet retainers which are riveted to the bottom plane of the box. I connected the heater elements through over-temperature protection switches that were in stock electric heaters. I run a 2.5mm2 wires cable through plastic cable-through holder in the same place where one of water pipes is going out on stock liquid heater. This way the heater cable goes out into the engine bay where the liquid heating pipe used to come in. Here is box view from top.

Must admit that initially box came a bit over-dimensioned because stock radiator had round corners while mine had square ones. So to fit it in I had to grind corners a bit and adjust bends.  After that it lost a bit of it's tidy look. Here is the view from bottom.

I then placed the newly made electric heater box into ventilation box. Had to make a bit wider openings in ventilation box plastic where white cable holder goes out.

Then I've assembled back the ventilation box and ran a short test by connecting the fan to 12V battery and heater elements to 220V socket. After few brief moments I had around 50 degree Celsius hot air  blowing out - should be ok in a car on a cold day. Of course car's nominal voltage is around 140V so out of two 1500W@220V AC rated each heater elements I should be getting around 1900W. I also placed a small DS18B20 temperatures sensor into the box airflow path to be able to measure the air temperature and regulate the heater element temperature from a control box BCMS. 

Then I put assembled ventilation box back into the car and started assembling the cabin dashboard again.

This dashboard disassembling and assembling served another purpose too - wiring the additional cables. Cabling took quite some time to do as I was only taking out unneeded cables that were going to ICE and leaving the useful ones. I also had to wire some additional cables to be able to control RPM, fuel level and temperature gauges of Honda's stock dashboard instruments from BCMS. The whole process took about a week working on evenings after the work.


Real characteristics of Kelly KP Series 0-5V Throttle

As I made the control box and most of its wiring I started experimenting with it on the the desk with 1kW lights connected as the load instead of the motor. I found that I can control the lights with the throttle quite ok but sometimes they go off and sometimes they don't when I release the throttle. I thought that it might be Kelly controller glitches as I didn't trust it.
But then I looked at the voltage of throttle output signal through oscilloscope. And I didn't like what I saw - at released throttle was sometimes producing rippled voltage of sawtooth shape oscillating between 0.9 and 1.2 V at around 30Hz frequency. Sometimes it was stable at 0.9V. I suspected that 5V supply to the pedal could be noisy causing hall chip to generate this ripple but when analyzed it with oscilloscope I found it pretty normal 5V power supply noise ripple of only around 0.1mV which should not be an issue at all. Anyway I added extra 0.1uF capacitor to measured the 5V supply voltage which didn't change anything - I was still seeing the nasty ripple on released throttle. In addition to that the movement range from low throttle to the moment when the output voltage started rising was quite big - around 6 degrees out of ~35 degree range. I also noticed that full open throttle was giving only around 3.8V. This is certainly not even close to 5V stated by Kelly.
After finding that I contemplated on two options: send the throttle back to Kelly for repair or repair it myself. First option looks logical when you are dealing with respectable company in your country which will provide good customer support to resolve the problem. With Kelly it is different story - I've read many posts about terrible support of Kelly in the forums and it could be likely that I would spend additional money and time for shipments of repair/replacement with likely the same end result - throttle not working in full range. It would also be little use trying to claim the money back as Kelly is in China where business ethics in some companies could be close to jungle and their government does nothing to protect foreign customers. So I decided to look what's inside and correct the throttle box myself.
I opened the box using Dremel-style grinder cutting through black potty on the bottom to get to the bottom cover. Finally I opened the bottom cover. I found crudely hand-made marginal quality mechanics that is typical for Chinese products with magnet rotating against hall sensor. The hall sensor is AH49E which is produced by several manufacturers with similar characteristics. Datasheet could be found on http://www.bcdsemi.com/upload/datasheet/AH49E%20P1.1%2020080619.pdf. Datasheet shows that this device has output operating voltages range from around 0.9V to 4.2V so it can never output the voltages close to 0-5V range. This confirms that trying to send the throttle for repair would be waste of time and money as they would likely be replacing it with another AH49E. Replacement could possibly fix auto-generation sawtooth ripple on the output but not the total range.
I understand that the range of 0.9 to 3.8V could also work by adjusting the throttle lower and upper dead limits but such solution was not appealing to me at all.
So I replaced the AH49E with Melexis MLX90215 programmable hall effect sensor. I calibrated the sensor's gain and offset by programming it's parameters and I got stable voltage in 0.2V-4.8V range for full throttle movement range from moment when throttle switch clicks from idle till the throttle is full open. This is what I would expect from product which states that it's range is 0-5V

Conclusion: Kelly KP Series 0-5V Throttle can only give you voltage ranges from 0.9V to 4.2V in best case if you are lucky getting the one which was somehow adjusted to give these voltages range over full throttle movement. But in reality I would expect the range of usable movement and voltages to be even worse. In addition I am not sure self-generation of ripple would not be seen on other units as well. If I was a manufacturer I would not dare to put such specifications for cheating customers. I wish I read something like that before buying anything from Kelly.


BMS Charging System Works!

I've been working intensively on my battery modules, BMS and charger to get them working together. So here's the story in sequence.

Fun with SMD microbs

My battery modules are made from small SMD components which I was never working with before. My previous electronics experience was with bigger sized and DIP components. So it was a new learning experience as I had with many other areas on this project - a lot of fun and satisfying moments.
On the first PCB I tried soldering these components with fine tip soldering iron which worked but I wanted to try a different way. Popular way which is also used in industry is placing components on the PCB with SMD soldering paste and treating it with heat at which solder melts. In serial manufacturing the boards are soldered by putting them in special owens.
Alternative for manual soldering is soldering with hot air. I used the hot air way although owen method should work too if had an owen. In the beginning I used some small Portasol gas soldering iron with air blower tip. It worked fine but I was not happy that I cannot control the temperature of the air and I can only have a rough estimate of this temperature. So I bought a hot air soldering station which I know I'll use in future too. Soldering the boards with it was an easy job. The tedious but not difficult part of the process was putting SMD paste and placing the components on the boards before blowing them with hot air. No problem with shaky hands and sight. Surprisingly to myself I found it easiest to do it without any magnifying goggles although I had them. My wife said I could be a surgeon. I guess I could :-)
Anyway I liked the process and I think it would have taken me notably longer with traditional components with all legs bending, sticking, soldering and crimping. It has taken me around 2-3 hours for soldering each board consisting of 10 battery modules. So manufacturing 45 modules was not bad at all.
Here is one batch of 10 manufactured boards

And here is a massive array of 36 as 9 others were already sitting on the cells as I was taking this photo :)

Battery module features

I'd like to list features that my designed battery modules have. It was quite long and iterative design and development process and I am happy I finished it. I like the final result.

Here is the list of battery module version 1 features:
- Balancing type: shunting current controlled by microcontroller
- Microcontroller: ATMEL ATtiny25V
- Operating voltage range: 2...5V
- Shunting resistance: 2.35Ohm 10 Watt
- Shunting current control: Pulse Width Modulation
- Communication protocol: custom serial protocol with LIN style sync frame
- Comm. bit rate: 2400bps (may work up to 9600bps or more)
- Comm. line: 1 wire going from module to module in chain - very easy and tidy installation
- Communication isolation: 1 Optocoupler on first module on the 2-wire input from BMS and 1 on the output to BMS over 2 wires back
- Measured parameters: Cell voltage, Cell temperature and shunting PWM value
- Voltage measurement accuracy: +/- 0.01V
- Temperature measurement accuracy: +/- 5C
- Parameter read commands: Read Voltages from all cells, Read Voltage from specific cell, Read Temperatures from all cells, Read Temperature from specific cell, Read PWM values from all cells, Read PWM value from specific cell
- Parameter write commands: Write balancing voltage to all cells, Write balancing voltage to specific cell, Write temperature calibration to all cells, Write temperature calibration to specific cell, Write max allowed temperature to all cells, Write max allowed temperature to specific cell, Write max allowed PWM to all cells, Write max allowed PWM to specific cell, Write cell IDs to all cells and read back the cell count.
- Automatic cell enumeration and counting via Write cell IDs command from BMS
- Automatic microcontroller sleep on idle (no comm from BMS for some time - like 60 seconds) to save power
- Automatic shunting current control via PWM to keep preset balancing voltage
- In-circuit microcontroller reprogamming capability via 1-wire debugWire interface
- LED indication of communication activity and PWM balancing mode
- Voltage spikes protection of the circuit supply, input and output signals
- Low cost due to carefully selected components and their count on each module

BCMS prototype board upgrade with CAN bus and Real-Time clock

I've added Microchip's MCP2515 CAN controller with Philips PCA82C260 CAN interface to be able to communicate with my new PFC 3kW charger from www.hztiecheng.com. My ATmega640 talks to MCP2515 over SPI interface. After few evenings of programming I made CAN bus work and my BCMS started talking to the charger. I can set charger voltage and current and enable charging. Charger returns me actual voltage and current plus the status with info such as overtemperature or battery fault flags. The charger model I have is named 144V/16A but as I learned it is capable of producing currents up to 20A.
Then I worked on programming the charging algorithm which would be suitable for my battery pack. It was done quite quickly to the point where I needed to have some reliable timekeeping. So I decided to finaly implement the real time clock (RTC) into my BCMS.
The RTC is made on Maxim DS1307 chip which is talking to my main processor over I2C interface. It has a Lithium backup button battery to keep the clock running when all the batteries are disconnected.
Here is the photo of the board with the processor off

And here is with the processor on

BMS charging demo

Finally here is a short movie demonstrating my BMS working with charger to charge and balance the battery pack

At this point I will be starting to put all electronics into control box, test it and then put into the car and connect to the motor. I'll put battery boxes with batteries in the back trunk first to have test drives and tune the system. Later I'll put them under car's belly as intended.


Battery Boxes Nearly Finished; PCBs, Components and Charger Arrived

I've done several things since last update. None of them are finished yet but the goal is getting closer and closer.
Metal works on battery boxes are finished. Boxes and their metal retainer parts are welded, grinded and sanded to the final look. Here is the photo of them stacked together as they would be placed under the car. Only plastic battery retainers and covers left to do before putting the batteries in.

I received manufactured PCBs of my battery modules. The quality is good and boards look nice.

I placed them on the batteries in one box to see how it would look just for fun.

I also received a package of ordered SMD components for my modules - almost 800 items. It would be interesting experience to manufacture 45 boards :) So far I had time just to solder essential components to start the module working. You can see the LED lit on the center of the board that is controlled by ATtiny25V microcontroller which is placed on the center of other side of PCB. It now just runs basic test controlling the shunting current running through two big white 5W resistors. It controls the current using PWM signal switching the big black MOSFET. With 4V voltage shunting current can reach up to 1,7A and resistors get so hot that it is just bearable to touch them with the finger. With this voltage they can dissipate up to 7W of power but I think I'll use about half of that and will have longer balancing stage when charging.

I also received the charger from China. I like the quality and the look. I like that it is sealed it does not have any fans - no worries about moisture and dust getting inside. It came with AC socket and European AC cable. It has couple of meters of thick cable that should go to the batteries on the other side. It also came with CAN bus interface adapter - small black box with white label. I asked for this model because my BCMS will be able to control the charging voltage and current through all charging stages and also receive the status from the charger - cool. Of course I'll have to implement the CAN interface electronics and software of my BCMS. But that is worth it as at the end I will have a well-integrated flexible BMS system.

I also started discussion with local transport inspection and transport experts company regarding registration of my EV to make it street legal. They said they don't have any paperwork prepared yet and I would be the first to do it in Lithuania. So I'll have to work with them to prepare the technical requirements for registration and then arrange my EV inspection according to these requirements. It means more paperwork for me and I would need to push the process forward but I will be able to influence the requirements to better suit my own needs and possibly make them good for other EV converters here. So I reckon it's not that bad.
There are many things waiting to be done but step by step I'm slowly doing them to get closer to the goal.


BCMS and Battery Boxes

I caught some nasty flu on Thursday and needed to sit at home during almost whole weekend. Luckily wife with baby were out during it - visiting parents, so they were safe. Once I was feeling better I resumed some works on electronics and programming. I sent an order for battery module 45 PCBs. I expect to have them on Friday.

I've started programming the PDA graphical user interface. I'll use IPAQ or similar PDA with PocketPC 2003 system which will be mounted in the dashboard and give all information about car, batteries, temperatures, etc. For starters I just made it read the serial output sentences which are generated by BCMS. The protocol is NMEA which is widely used in GPS devices. I've created some custom sentence names which are used in my BCMS. The idea is that I'll have all parameters information along with GPS position of the car which will help car performance analysis after. Below is PDA emulator's program's simple screen which reads the data from serial port and displays in simple text fields. The parameters displayed are actually measured by BCMS.

When I got better on Sunday afternoon I went to garage and made a short session working on battery boxes. I welded the top mounting flange with one hole for initial mounting of the box. I also welded a M8 bolt part at the botom of the box where the batteries are being put. This bolt will be used to secure and press the batteries with V-shaped frame shown below. This will prevent any possible swelling of the cells.

Next will be a lot of grinding of the boxes to remove excess metal from the welds and improve aesthetic view. Nasty part will be cleaning the welds in inner corners of the boxes as it is not yet done properly. Then I will put stainless steel sheet into the boxes walls and they will be almost finished. Almost... There will be plastic retainer bars inside to stop cells from bouncing in the box. And also some plastic or maybe metal sheet cover on top of the boxes with openings at both ends to allow access for boxes interconnections.


Battery boxes and Control box

Battery boxes

As mentioned earlier I decided to place all the batteries underneath the car where the fuel tank used to be. After making measurements thinking and modeling I chose to make battery boxes to be fitted from underneath the car thus preserving most of the body structure and not weakening it with serious modifications. The reason is that I am thinking ahead about how I would present the car to inspectors who would be checking it to make street legal. Less modifications to the frame - less problems and questions asked.
So, I will make batteries trays/boxes fit and secure from underneath making only few openings from inside to complete the wiring once the boxes are placed in. There will be 5 battery boxes each holding 9 cells. Each box gross weight will be approx 32kg so it will be manageable to lift it to the car from underneath and secure it by hands of two persons. So you will be able put the whole 160kg of batteries box by box.
Here is the sketch of the battery boxes made with Google SketchUp:

Each box is a frame of stainless steel angle 20x20x3 with 0.8mm stainless steel sheet walls protecting the batteries from outside elements like water, sand and stones (remember they will be hanging under car's belly).
The production of boxes has started. I have made the welding jig to make their geometry solid. Without it would be very difficult to weld the steel boxes with decent precision and correct geometry.

Here is one of the boxes fit test with 9 ThunderSky cells. The cells have their contacts covered with tape - you don't want to short circuit expensive cell, don't you?

Control box

Control box is designed to hold all electronic equipment inside and provide a heat sink for controller and down converter. First of all I don't want any sensitive electronics or high voltage circuit hang outside where the could be touched by water, dirt or anyone's non-careful hands. Safety first. For example a session in car wash could lead to serious components damage or even fire in unfortunate circumstances. Therefore I decided to put all such pieces in one safe place - control box.
The control box will hold Kelly controller, controller cooling fans, DC 144V/13.8V down-converter, main contactor, circuit breaker, shunt resistor, throttle hall converter, multiple 12V control relays and fuses and BCMS master board. It will have two thick cables coming in from batteries, two going to the motor and low voltage signal cables going to many places.
The box is made from 10mm aluminium plate which makes base, 3 walls and internal spacer. The plates are bolted together by M5 bolts. The cover will be from 0.8mm stainless steel sheet.
The inside is divided by internal spacer wall into two sections: high voltage/power section and low voltage section. The sketch is shown in picture. There are two big black fans on top which are placed on the opposite side of the wall where the motor controller is placed for maximum heat dissipation effectivenes. The black box on the opposite side of the controller is DC down-converter. Both controller and converter terminals are placed so that they go out to the high voltage connections subsection on the upper right side of the box. This subsection is separated by plastic spacer plate which isolates the connections subsection to prevent any water or dirt entering there. Contactor, shunt and fuse are shown in this subsection. The subsection below contains BCMS board on the left and throttle pot on the right. Throttle handle is attached on the outside of the box. There will be gas cable connected to it.
The box will be put in the car above the motor in approximate location shown in sketch below. I hope you have enough imagination to see the yellow front of the car, black tire, red motor below and gray brakes cylinder assembly :).

Production of the box has started. The aluminium plates were cut (believe me it is daunting when you don't have the disk saw powerfull enough). A lot of drilling and threading was made.

Some power wiring was made with thick copper bars covered in yellow heat-shrink tubing. Production of each bar to about 1 hour - it looks small but eats your time very quickly.

A bit closer view. Note the two big holes drilled in the base - that's where the thick cables will go to the motor.

That's it for now. Some progress, not too much as I am really struggling to find time to sneak into garage and do some work on my HR-EV.